Upload
phungdien
View
214
Download
2
Embed Size (px)
Citation preview
69
RING-FENCING*
STEVEN L. SCHWARCZ†
ABSTRACT
“Ring-fencing” is often touted as a regulatory solution to problems in
banking, finance, public utilities, and insurance. However, both the precise
meaning of ring-fencing, as well as the nature of the problems that ring-
fencing regulation purports to solve, are ill-defined. This Article examines
the functions and conceptual foundations of ring-fencing. In a regulatory
context, the term can best be understood as legally deconstructing a firm in
order to more optimally reallocate and reduce risk. So utilized, ring-
fencing can help to protect certain publicly beneficial activities performed
by private-sector firms, as well as to mitigate systemic risk and the too-big-
to-fail problem inherent in large financial institutions. If not structured
carefully, however, ring-fencing can inadvertently undermine efficiency
and externalize costs.
TABLE OF CONTENTS
I. INTRODUCTION ................................................................................... 70
II. DEFINING RING-FENCING ............................................................... 72
A. RING-FENCING TO MAKE A FIRM BANKRUPTCY REMOTE ......... 74
B. RING-FENCING TO HELP A FIRM OPERATE ON A
STANDALONE BASIS ................................................................... 77
* Copyright © 2014 by Steven L. Schwarcz.
† Stanley A. Star Professor of Law & Business, Duke University School of Law, and Founding
Director, Duke Global Capital Markets Center; [email protected]. I thank Iman Anabtawi,
Emilios Avgouleas, Lawrence G. Baxter, Andromachi Georgosouli, and participants in faculty
workshops at Queen Mary, University of London, and Campbell University School of Law for helpful
comments and Seth M. Bloomfield, Arie Eernisse, and Min B. Lee for valuable research assistance. The
author has testified as a ring-fencing expert before the Maryland Public Service Commission in the
merger of Exelon Corporation and Constellation Energy Group. He also helped advise Sir John Vickers,
Chair, and the Secretariat of the U.K. Independent Commission on Banking in connection with that
Commission’s 2011 report on U.K. bank regulation (the “Vickers Report”), which proposes a form of
bank ring-fencing.
70 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
C. RING-FENCING TO PRESERVE A FIRM’S BUSINESS AND
ASSETS ........................................................................................ 77
D. RING-FENCING TO LIMIT A FIRM’S RISKY ACTIVITIES AND
INVESTMENTS ............................................................................. 78
1. The Vickers Report ............................................................... 78
2. The Glass-Steagall Act ......................................................... 79
3. The Volcker Rule .................................................................. 80
E. FUNCTIONAL DEFINITION ............................................................ 81
III. NORMATIVE ANALYSIS .................................................................. 83
A. RING-FENCING TO CORRECT MARKET FAILURES....................... 84
1. Monopolies and Other Forms of Noncompetitive
Markets ................................................................................. 85
2. The Public-Goods Problem ................................................... 86
3. Information Failure ............................................................... 89
4. Agency Failure...................................................................... 91
5. Responsibility Failure ........................................................... 93
B. RING-FENCING TO PROTECT AGAINST SYSTEMIC RISK .............. 95
1. Minimizing Panics ................................................................ 95
2. Creating Modularity.............................................................. 96
IV. COSTS AND BENEFITS .................................................................... 97
A. BANK RING-FENCING ................................................................. 98
1. The Glass-Steagall Act ......................................................... 98
2. The Vickers Report ............................................................. 101
B. UTILITY RING-FENCING ............................................................ 105
C. RING-FENCING OF SIFIS............................................................ 106
1. Protecting the Publicly Beneficial Activities Performed
by SIFIs .............................................................................. 106
2. Protecting Against the Systemic Failure of SIFIs ............... 107
V. CONCLUSION .................................................................................... 108
I. INTRODUCTION
“Ring-fencing” is often touted as a potential regulatory solution to
problems in banking, finance, public utilities, and insurance.1 A prominent
U.K. government report proposes ring-fencing banks by legally separating
certain of their risky assets from their retail banking operations.2 Federal
1. Ring-fencing (ring-fence) is also sometimes referred to as “ringfencing” (“ringfence”).
2. This is the principal recommendation of the Vickers Report. INDEP. COMM’N ON BANKING,
FINAL REPORT: RECOMMENDATIONS 9–12 (2011), available at
http://www.ecgi.org/documents/icb_final_report_12sep2011.pdf [hereinafter VICKERS REPORT].
2013] RING-FENCING 71
regulators in the United States are considering requiring the ring-fencing of
systemically important financial institutions, including banks, to reduce
systemic risk.3 They also are attempting to implement the so-called
“Volcker Rule,” a form of ring-fencing.4 Congress has been considering
enacting a ring-fencing scheme proposed in federal “covered bond”
legislation,5 which would parallel European ring-fencing of certain secured
transactions.6 State regulators often require the ring-fencing of utility
companies by legally separating their risky assets and operations from the
public-utility function.7 And the leading insurance standard-setting and
regulatory support organization in the United States is proposing the
increased ring-fencing of insurance companies.8
Because it is proposed in different contexts as a solution to ostensibly
3. Federal Reserve Governor Daniel K. Tarullo has proposed ring-fencing the U.S. operations
of large foreign banks and of systemically important financial institutions. See Daniel K. Tarullo,
Member, Bd. of Governors of the Fed. Reserve Sys., Speech at the Yale School of Management
Leaders Forum (Nov. 28, 2012), available at http://www.federalreserve.gov/newsevents/speech/
tarullo20121128a.htm; Jonathan Spicer, Update 2-Fed’s Tarullo Urges Global Action on Regulating
Banks, REUTERS (Feb. 22, 2013), http://www.reuters.com/article/2013/02/23/usa-fed-tarullo-regulation-
idUSL1N0BMDRM20130223.
4. As of September 2013, the Volcker Rule has yet to be finalized for implementation. Drawing
Bright Lines for Banks, BLOOMBERG (Sept. 3, 2013, 6:00 AM), http://www.bloomberg.com/news/2013-
09-03/drawing-bright-lines-for-banks.html; Cheyenne Hopkins, Dodd-Frank Implementation Defended
by U.S. Regulators, BLOOMBERG (Feb. 14, 2013, 8:41 AM), http://www.bloomberg.com/news/2013-02-
14/dodd-frank-implementation-defended-by-u-s-regulators.html.
5. On July 22, 2010, Rep. Scott Garrett (R-NJ) and cosponsors Rep. Paul E. Kanjorski (D-PA)
and Rep. Spencer Bachus (R-AL) introduced the “United States Covered Bond Act of 2010” (H.R.
4884, later renumbered as H.R. 5823, 111th Cong.). This bill has been recommended by the House
Committee on Financial Services for consideration by the full U.S. House of Representatives. United
States Covered Bond Act of 2010, H.R. 5823, 111th Cong. (2010), available at
http://www.govtrack.us/congress/bills/111/hr5823. The bill was reintroduced on March 8, 2011, by
Rep. Scott Garrett (R-NJ) and cosponsor Rep. Carolyn Maloney (D-NJ) as the “United States Covered
Bond Act of 2011” (H.R. 940, 112th Cong.). United States Covered Bond Act of 2011, H.R. 940, 112th
Cong. (2011), available at http://www.govtrack.us/congress/bills/112/hr940. The House Financial
Services Committee voted 44–7 in favor of the bill, but it has not yet been enacted by the full U.S.
House of Representatives. Jon Prior, House Committee Clears Framework for Covered Bonds,
HOUSINGWIRE (June 22, 2011, 4:45 PM), http://www.housingwire.com/news/2011/06/22/house-
committee-clears-framework-covered-bonds. For a discussion of covered bonds, see infra Part II.A.
6. Steven L. Schwarcz, The Conundrum of Covered Bonds, 66 BUS. LAW. 561, 565–68 (2011).
7. See, e.g., Charles E. Peterson & Elizabeth M. Brereton, UTAH STATE DEP’T OF COMMERCE,
REPORT ON RING-FENCING 35–39 (2005), available at http://www.psc.utah.gov/utilities/telecom/
05docs/0505301/Dir%20Test%20C%20Peterson%20DPU%20Exhibit%2010.1.doc (summarizing
selected state laws that require the ring-fencing of public utility companies).
8. See Insurance Oversight and Legislative Proposals: Hearing Before the Subcomm. on Ins.,
Hous., & Cmty. Opportunity of the H. Comm. on Fin. Servs., 112th Cong. 54–57 (2011) (statement of
Daniel Schwarcz, Assoc. Professor, Univ. of Minn. Law Sch.) (critiquing a proposal by the National
Association of Insurance Commissioners (NAIC) for a “windows and walls” approach to insurance
group regulatory supervision).
72 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
different problems, ring-fencing is inconsistently defined; and even within
a given context, it is often ill-defined. Part II of this Article attempts to
define ring-fencing by examining its functions. That examination shows
that ring-fencing can best be understood as legally deconstructing a firm in
order to more optimally reallocate and reduce risk. The deconstruction can
occur in various ways: by separating risky assets from the firm, by
preventing the firm itself from engaging in risky activities or investing in
risky assets, or by protecting the firm from affiliate and bankruptcy risks.
This increased definitional clarity raises important normative
questions about when and how ring-fencing should be used as an economic
regulatory tool. Which firms, for example, should be subject to ring-
fencing? Which “risky” assets should be separated from the firm, and how
should that separation occur? Which “risky” activities and asset
investments should the firm not engage in, and how should that
engagement be prevented? Which affiliate “risks” should the firm be
protected from, and how should that protection be implemented?9 Part III
of the article attempts to answer these questions.10
Ring-fencing, however, can also impose costs, potentially
undermining efficiency. Part IV of the article critiques actual and proposed
regulatory uses of ring-fencing in light of their potential costs and benefits.
II. DEFINING RING-FENCING
Any attempt to define ring-fencing faces a threshold question: How
should a financial regulatory concept be defined?11 In answering this
9. Although the transferring of assets to offshore accounts to avoid liability is, for example,
colloquially called ring-fencing, Ring Fence, INVESTOPEDIA, http://www.investopedia.com/terms/r/
ringfence.asp#axzz2M3mADa00 (last visited Oct. 28, 2013), such a practice may better be described as
a form of judgment proofing. Cf. infra text accompanying notes 92–95 (comparing ring-fencing and
judgment proofing).
10. In Part III, the Article examines, for example, ring-fencing used to help protect certain
publicly beneficial activities that are performed by private-sector firms, such as utility companies and
banks. This is the purpose of ring-fencing used to protect essential public utility services and, under the
Vickers Report, proposed to protect retail banking services. See infra text accompanying notes 224–29,
248–50. It is also one of the purposes of ring-fencing used in securitization and covered bond
transactions. See infra note 204 and accompanying text. The Article also examines ring-fencing used to
help mitigate systemic risk and the too-big-to-fail problem inherent in large banks and other financial
institutions. This is the purpose of ring-fencing proposed under the Dodd-Frank Act for systemically
important financial institutions. See infra text accompanying notes 237–40, 263–65.
11. Ring-fencing is clearly a financial regulatory concept when used for banks and other
financial institutions, the uses on which this Article primarily focuses. Ring-fencing is less clearly a
financial regulatory concept when used for public utility companies and insurance companies. This
Article only incidentally focuses on ring-fencing insurance companies. Although the Article provides
greater focus on ring-fencing utilities, it uses utility ring-fencing to draw an analogy between a utility
2013] RING-FENCING 73
question, one confronts “the lack of an agreed upon methodology on how
to . . . define legal concepts.”12
Financial regulation governs how law regulates financial players, such
as banks and other financial institutions. It thus is not an abstraction; there
are real economic consequences. Even a normative definition of a financial
regulatory concept should therefore be rooted pragmatically, taking into
account how, functionally, the concept is used in the real world.13 This
functional approach would avoid the “misunderstanding and unwanted
interpretations”14 that can result by defining a concept in a new way. This
approach also acknowledges that “[i]f all concerned people understand
concepts A, B and C in a specific way due to their foundation
in . . . common practice, it is preferable to use them rather than the more
abstract concept of D that contains A, B and C.”15
Being a financial regulatory concept, ring-fencing should likewise be
defined functionally, taking into account its real-world use. Perhaps the
most common function of ring-fencing is to protect a firm from becoming
subject to liabilities and other risks associated with bankruptcy.16 This is
usually called making the firm “bankruptcy remote.”17 Another function of
ring-fencing is to help ensure that a firm is able to operate on a standalone
basis even if its affiliated firms fail.18 Yet another function of ring-fencing
is to protect a firm from being taken advantage of by its affiliated firms—
company providing publicly beneficial utility services and a bank or other financial institution
providing publicly beneficial financial services. In drawing that analogy, the Article distinguishes
differences between utility companies, on the one hand, and banks and other financial institutions, on
the other hand, that could impair the analogy. See, e.g., infra text accompanying notes 251–53
(explaining how differences in the ring-fencing of those entities result from differences in those entities’
characteristics).
12. Lorenz Kähler, The Influence of Normative Reasons on the Formation of Legal Concepts, in
CONCEPTS IN LAW 81, 90 (Jaap C. Hage & Dietmar von der Pfordten eds., 2009) (footnote omitted)
(citing Dennis M. Patterson, Dworkin on the Semantics of Legal and Political Concepts, 26 OXFORD J.
LEGAL STUD. 545, 553 (2006)).
13. “Indeed, a normative definition should strive to achieve an optimal regulatory or other
clarifying purpose, otherwise the definition is merely an academic exercise.” Steven L. Schwarcz, What
Is Securitization? And for What Purpose?, 85 S. CAL. L. REV. 1283, 1289–90 (2012) (footnote omitted)
(examining how, normatively, to define the financial concept of securitization).
14. Kähler, supra note 12, at 86.
15. Id.
16. See infra Part II.A (discussing this function of ring-fencing). Cf. Schwarcz, supra note 6, at
567 (discussing ring-fencing in structured covered-bond regimes).
17. S.L. Schwarcz, Securitization and Structured Finance, in HANDBOOK OF KEY GLOBAL
FINANCIAL MARKETS, INSTITUTIONS AND INFRASTRUCTURE 565, 567 (Gerard Caprio Jr. et al. eds.,
2013).
18. See infra Part II.B (discussing this function of ring-fencing).
74 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
essentially preserving the business and assets of the ring-fenced firm.19
And still another function of ring-fencing is to limit a firm’s risky activities
and investments.20
The discussion below examines and provides examples of these
functions. The examples focus on ring-fencing as a form of financial
regulation. That use of ring-fencing should be—and at the end of Part II.E,
is—distinguished from judgment proofing, a superficially related but
diametrically opposed concept.21
A. RING-FENCING TO MAKE A FIRM BANKRUPTCY REMOTE
Ring-fencing can be, and often is, used to make a firm bankruptcy
remote.22 This use of ring-fencing is most common in securitization and
covered bond transactions. It also is common for public utility companies,
which are private-sector companies that generate or otherwise provide the
public with power, clean water, communications, and other essential
utilities.23
In securitization and covered bond transactions, the ring-fenced firm is
normally a special purpose entity (“SPE”) acting on behalf of an affiliated
firm that wants to raise financing. Bankruptcy remoteness enhances the
creditworthiness of the SPE, thereby enabling it to issue securities to
investors at lower cost, and in a manner that more efficiently allocates risk,
than if the affiliated firm issued the securities.24 Ring-fencing is also
commonly used to make utility companies bankruptcy remote. This use of
ring-fencing is a response to holding company structures, in which the
utility company is often a subsidiary of one or more operating companies
that may engage in riskier transactions. Bankruptcy remoteness helps to
ensure that the utility company can continue providing essential utilities to
the public, notwithstanding the bankruptcy of the parent company.25
Ring-fencing can achieve bankruptcy remoteness contractually or,
where appropriate legislation exists, by legislative fiat.26 Securitization
transactions typically are ring-fenced contractually to achieve bankruptcy
19. See infra Part II.C (discussing this function of ring-fencing).
20. See infra Part II.D (discussing this function of ring-fencing).
21. See infra text accompanying notes 92–95 (explaining that distinction).
22. Cf. supra text accompanying notes 16–17 (defining bankruptcy remoteness).
23. References in this Article to utilities or utility companies hereinafter will mean public utility
companies.
24. For an efficiency analysis of this risk allocation, see Steven L. Schwarcz, Securitization Post-
Enron, 25 CARDOZO L. REV. 1539, 1553–69 (2004).
25. See infra text accompanying notes 36–42.
26. Schwarcz, supra note 6, at 567.
2013] RING-FENCING 75
remoteness.27 This includes protecting the SPE from both voluntary and
involuntary bankruptcy. The former is achieved through corporate
governance techniques that limit the ability of the SPE’s managers to file
for bankruptcy.28 The latter is achieved by limiting the SPE’s ability to
incur other-than-specified indebtedness.29 These steps also include
protecting the SPE from equitable and other corporate veil-piercing threats,
such as “substantive consolidation.”30 That typically is achieved by
requiring the SPE to maintain strict arm’s length formalities with its
affiliates.31
Covered bond transactions are ring-fenced either legislatively, in
jurisdictions that have enacted covered bond statutes, or contractually in
other jurisdictions.32 The steps needed to contractually ring-fence covered
bond transactions can parallel the ring-fencing steps taken in securitization
transactions,33 although there are some notable differences.34 In both cases,
however, the goal is to make the covered bond transaction bankruptcy
remote.35
Utility companies are ring-fenced, to achieve bankruptcy remoteness,
through a combination of contract and legislation.36 Utilities are normally
operated in the United States, for example, through a holding company
structure, in which a parent company owns the shares of the utility
subsidiary.37 This structure provides greater flexibility because the parent is
not necessarily regulated as a utility, thereby enabling the corporate group
to raise capital on more favorable terms and to attract and cultivate a larger
27. Id.
28. STEVEN L. SCHWARCZ, STRUCTURED FINANCE: A GUIDE TO THE PRINCIPLES OF ASSET
SECURITIZATION § 3:2.1 (3d ed. 2003) [hereinafter STRUCTURED FINANCE]. For example, the SPE’s
organization documents will require one or more of its managers to be independent of affiliated
companies. Id.
29. Id. § 3:3.
30. Id. § 3:4.
31. Id.
32. Schwarcz, supra note 6, at 567, 571.
33. Id.
34. For example, securitization is nonrecourse financing and covered bonds have full recourse to
the issuer. Id. Additionally, in a securitization transaction the transferred assets are treated as off the
originator’s balance sheet, while in a covered bond transaction the assets typically remain on the
issuer’s balance sheet. Id.
35. Id.
36. See Peterson & Brereton, supra note 7, at 35–39 (summarizing the legislation of Maryland,
Wisconsin, Virginia, Oregon, and New Jersey that uses ring-fencing techniques to achieve bankruptcy
remoteness for utility companies).
37. Richard D. Cudahy & William D. Henderson, From Insull to Enron: Corporate
(Re)Regulation After the Rise and Fall of Two Energy Icons, 26 ENERGY L.J. 35, 57 (2005).
76 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
pool of engineering talent.38 Nonetheless, as holding companies
increasingly have diversified their investments to riskier (nonutility) assets,
failures have increased.39 The resulting parent-company bankruptcies have
exposed the utility-subsidiaries to bankruptcy.40 To mitigate this risk,
utilities typically are operated as bankruptcy-remote subsidiaries of their
holding companies.41 The terms of such bankruptcy remoteness, including
the contractual means for achieving it, are usually mandated by the utility’s
regulator—in the United States, state public utility commissions.42
In 1997, for example, Enron acquired Portland General Electric
(“PGE”), which was regulated by the Oregon Public Utility Commission
(“OPUC”).43 The merger between Enron and PGE was contingent upon
terms stipulated by the OPUC,44 which (among other things) mandated that
PGE be held by Enron in a bankruptcy-remote structure.45 When Enron
eventually filed for bankruptcy, these ring-fencing measures protected PGE
from bankruptcy.46
The discussion above has illustrated how ring-fencing is commonly
used to achieve bankruptcy remoteness for utilities and in securitization
and covered bond transactions. Although it has other bankruptcy-remote
applications, ring-fencing is not typically used to achieve bankruptcy
remoteness in banking or insurance. The reason is path dependent: at least
in the United States, banks and insurance companies have not historically
been subject to bankruptcy law.47
38. Id.
39. Fred Grygiel & John Garvey, Fencing in the Regulated Utilities, PUB. UTIL. FORT., Aug.
2004, at 32, 32.
40. See, e.g., PUB. CITIZEN, CHANGES IN THE FINANCIAL HEALTH OF THE ELECTRIC AND
NATURAL GAS UTILITY INDUSTRIES SINCE THE PUHCA HEARINGS OF 2001 (2004), available at
http://www.citizen.org/documents/industryhealth.pdf (discussing the wave of bankruptcies that resulted
from PUHCA-exempt or “non-utility” businesses in 2003).
41. Grygiel & Garvey, supra note 39, at 32.
42. Id.
43. Peterson & Brereton, supra note 7, at 2, 13 (recommending the use of ring-fencing in Utah
and discussing the successful use of ring-fencing by the state of Oregon in the case of Portland General
Electric).
44. Id. at 13.
45. Id. at 15.
46. MILES H. MITCHELL ET AL., MD. PUB. SERVS. COMM’N, COMMISSION STAFF ANALYSIS OF
RING-FENCING MEASURES FOR INVESTOR-OWNED ELECTRIC AND GAS UTILITIES 14 (2005), available
at http://webapp.psc.state.md.us/Intranet/Reports/RevisedRing-FencingReport.pdf (recommending the
use of ring-fencing in Maryland and discussing the successful use of ring-fencing by the state of Oregon
in the case of Portland General Electric).
47. See, e.g., 11 U.S.C. § 109(b)(2) (2012) (excluding deposit-taking banks and domestic
insurance companies from federal bankruptcy law).
2013] RING-FENCING 77
B. RING-FENCING TO HELP A FIRM OPERATE ON A STANDALONE BASIS
Ring-fencing can also be used to help ensure that the ring-fenced firm
is able to operate on a standalone basis even if its affiliated firms fail. Such
assurance would be needed if, for example, a utility company is dependent
on its affiliates for goods and services, such as raw materials or
administrative or operating services.48 This form of ring-fencing thus
would include putting into place back-up contracts with independent third
parties to provide any such needed goods and services.
In the case of PGE’s acquisition by Enron, for example,49 PGE was
ring-fenced to ensure that it owned or leased the assets used in its
business.50 And in the case of the acquisition of Baltimore Gas and Electric
Company (“BGE”) by Exelon Corporation, BGE was ring-fenced to ensure
that it would be able to operate on a standalone basis even if its affiliated
firms failed.51
C. RING-FENCING TO PRESERVE A FIRM’S BUSINESS AND ASSETS
Ring-fencing can also be used to protect the ring-fenced firm from
being taken advantage of by affiliated firms. In a utility-company context,
this may entail mandating that all transactions between the utility and its
affiliates be arm’s length. In the case of PGE’s acquisition by Enron, for
example, the merger terms stipulated by PGE’s regulator52 required, among
other things, that PGE was “required to maintain books and records
separate from Enron; to maintain separate accounts; to continue to hold all
of its assets in its own name; and to enter into transactions with Enron only
as permitted by federal and state regulators.”53 In the case of the acquisition
of BGE by Exelon Corporation, the merger terms also imposed restrictions
48. Rockland Electric Company and Pike County Light & Power Co., for example, each relies on
its parent company, Orange & Rockland Utilities, Inc., to provide administrative services such as
customer account management and customer service. See O&R at a Glance, ORANGE & ROCKLAND
UTILS., INC., http://www.oru.com/aboutoru/oruataglance/index.html (last visited Oct. 29, 2013). In
another example, the utility National Fuel Gas Distribution Corporation, which provides natural gas to
customers in New York and Pennsylvania, relies on subsidiaries of its parent company, National Fuel
Gas Company, for its supply of natural gas. Nat’l Fuel Gas Co., Annual Report (Form 10-K), at 6, 10
(Sept. 30, 2012).
49. See supra text accompanying notes 43–46.
50. MITCHELL ET AL., supra note 46, at 14.
51. In the Matter of the Merger of Exelon Corporation and Constellation Energy Group, Inc.:
Hearing on Case No. 9271 Before the Pub. Serv. Comm’n of Md. 9 (2011) (rebuttal testimony of Steven
L. Schwarcz) [hereinafter Rebuttal Testimony of Schwarcz]. Prior to the merger, BGE was owned by
Constellation Energy Group, Inc.
52. See supra text accompanying note 44 (discussing that stipulation).
53. Peterson & Brereton, supra note 7, at 14.
78 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
on the amount of dividend payments that BGE could pay to its new owner,
limiting such payment unless BGE would retain a specified minimum net
worth after paying the dividend.54
This form of ring-fencing is also commonly applicable to banks.
Regulation may require, for example, that all transactions between a bank
and its affiliates be arm’s length.55
D. RING-FENCING TO LIMIT A FIRM’S RISKY ACTIVITIES AND
INVESTMENTS
Ring-fencing can also be used to limit a firm from engaging in risky
activities and making risky investments. The ring-fencing of bank activities
under the Vickers Report and the Glass-Steagall Act,56 as well as the
Volcker Rule, exemplify this approach.
1. The Vickers Report
In June 2010, the United Kingdom created the Independent
Commission on Banking (the “Commission”) to consider structural and
nonstructural reforms to the U.K. banking sector with the goal of
promoting financial stability and competition.57 Chaired by Sir John
Vickers, the Commission published its final report (widely known as the
“Vickers Report”) in September 2011.58 The goals of the Commission were
threefold: to “reduce the probability and impact of systemic financial
crises,” to “maintain the efficient flow of credit to the real economy,” and
to “preserve the functioning of the payments system and guaranteed capital
certainty and liquidity for small savers.”59 To meet these goals, the Vickers
Report recommended a combination of “structural reform and enhanced
loss-absorbing capacity.”60
The structural reform would require the ring-fencing of U.K. “retail”
54. Rebuttal Testimony of Schwarcz, supra note 51, at 5–6.
55. See Federal Reserve Act § 23A, 12 U.S.C. § 371c (2012) (imposing restrictions on
transactions between a bank and its affiliates).
56. Banking Act of 1933, ch. 89, 48 Stat. 162 (codified as amended in scattered sections of 12
U.S.C.). Glass-Steagall refers to sections 16, 20, 21, and 32 of the Banking Act of 1933. Section 16 was
codified as 12 U.S.C. § 24. Section 20 was codified as 12 U.S.C. § 377. Section 21 was codified as 12
U.S.C. § 378(a)(1). Section 32 was codified as 12 U.S.C. § 78.
57. See VICKERS REPORT, supra note 2, at 19.
58. Id. at 19–20. Vickers was then the Warden of All Souls College, University of Oxford.
Professor Sir John Vickers, ALL SOULS COLL., UNIV. OF OXFORD, http://www.all-
souls.ox.ac.uk/people.php?personid=72 (last visited Oct. 29, 2013).
59. VICKERS REPORT, supra note 2, at 20.
60. Id.
2013] RING-FENCING 79
banking activities—defined as banking activities for individuals and small
and medium-sized enterprises.61 Banks would be required to take deposits
from, and provide overdrafts to, those individuals and enterprises through
separate subsidiaries that could not engage in activities that might expose
them to loss, such as trading book activities, purchasing loans or securities,
and derivatives trading.62 That restriction on activities that could result in
loss exemplifies ring-fencing’s ability to limit a firm’s risky activities and
investments.
The Vickers Report also made recommendations about what it called
the “height” of the ring-fence; these recommendations implicitly address
aspects of ring-fencing’s other functions. Thus, the recommendation that
each ring-fenced subsidiary should be a separate legal entity that adheres to
strict arm’s length formalities63 appears to provide a measure of bankruptcy
remoteness.64 The recommendation that each ring-fenced subsidiary should
meet certain regulatory requirements for capital, liquidity, and funding65
appears to enable such subsidiary to operate, if needed, on a standalone
basis.66 And the recommendations that each ring-fenced subsidiary should
only engage in arm’s length transactions with affiliates and should have a
majority of its directors, including the chair, be independent67 should help
to preserve the subsidiary’s business and assets.68
2. The Glass-Steagall Act
The Glass-Steagall Act was enacted in the United States as part of the
Banking Act of 1933, responding to the Great Depression.69 The Glass-
Steagall Act ring-fenced deposit-taking banks by prohibiting them from
engaging in the securities business, which was perceived as risky.70
61. Id. at 10–11.
62. Id. at 11. Activities related to the provision of payment services to customers in the European
Economic Area (“EEA”) would also be permitted in the ring-fenced entity. Id. The Vickers Report also
permits flexibility for a ring-fenced subsidiary to provide straightforward banking services to large
domestic nonfinancial companies. Id. at 12.
63. Id. at 66–72.
64. See supra Part II.A.
65. VICKERS REPORT, supra note 2, at 71.
66. See supra Part II.B.
67. VICKERS REPORT, supra note 2, at 72.
68. See supra Part II.C.
69. See supra note 56 and accompanying text.
70. WILLIAM D. JACKSON, CONG. RESEARCH SERV., GLASS-STEAGALL ACT: FACT SHEET
(1999). Thus, Section 20 of the Glass-Steagall Act prohibited Federal Reserve member banks from
affiliating with organizations “engaged principally in the issue, flotation, underwriting, public sale or
distribution at wholesale or retail or through syndicate participation of stocks, bonds, debentures, notes,
or other securities.” Id. (quoting 12 U.S.C. § 377). Likewise, Section 21 of the Glass-Steagall Act
80 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
The Glass-Steagall Act’s ring-fencing was repealed on November 12,
1999 by the passage of the Gramm-Leach-Bliley Act.71 Under the Gramm-
Leach-Bliley Act, deposit-taking banks were allowed to affiliate in a
holding company structure with investment banks and other securities
firms.72
3. The Volcker Rule
In response to the recent financial crisis, former Federal Reserve
Chairman Paul Volcker proposed that because bank deposits are federally
guaranteed,73 deposit-taking banks should be restricted from making risky
investments.74 This proposal became known as the “Volcker Rule.”75 The
substance of the Volcker Rule was implemented by the Dodd-Frank Wall
Street Reform and Consumer Protection Act,76 enacted in July 2010.77 In
relevant part, that Act prohibits banks from (1) “engag[ing] in proprietary
trading”78 or (2) “acquir[ing] or retain[ing] any equity, partnership, or other
ownership interest in or sponsor[ing] a hedge fund or a private equity
prohibited “securities firms from engaging in ‘the business of receiving deposits.’” Id. (quoting 12
U.S.C. § 378).
71. Gramm-Leach-Bliley Act, Pub. L. No. 106-102, 113 Stat. 1338 (1999), available at
http://www.gpo.gov/fdsys/pkg/PLAW-106publ102/html/PLAW-106publ102.htm.
72. See 12 U.S.C. § 1841(p) (2012) (defining a Financial Holding Company); S. COMM. ON
BANKING, HOUS., AND URBAN AFFAIRS, GRAMM-LEACH-BLILEY: SUMMARY OF PROVISIONS, available
at http://www.banking.senate.gov/conf/grmleach.htm.
73. Paul Volcker, Op-Ed., How to Reform Our Financial System, N.Y. TIMES (Jan. 30, 2010),
http://www.nytimes.com/2010/01/31/opinion/31volcker.html?pagewanted=all.
74. Id.
75. David Cho & Binyamin Appelbaum, Obama’s ‘Volcker Rule’ Shifts Power Away from
Geithner, WASH. POST (Jan. 22, 2010), http://www.washingtonpost.com/wp-
dyn/content/article/2010/01/21/AR2010012104935.html.
76. Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010, Pub. L. No. 111-203,
124 Stat. 1376 (2010).
77. William J. Sweet, Jr. & Brian D. Christiansen, The Volcker Rule, INSIGHTS (Skadden, Arps,
Slate, Meagher & Flom), July 9, 2010, available at http://www.skadden.com/newsletters/FSR_The_
Volcker_Rule.pdf.
78. 12 U.S.C. § 1851(a)(1)(A) (2012). “Proprietary trading” is defined as engaging as a principal for the trading account of the banking entity or [relevant] nonbank financial company . . . in any transaction to purchase or sell, or otherwise acquire or dispose of, any security, any derivative, any contract of sale of a commodity for future delivery, any option on any such security, derivative, or contract, or any other security or financial instrument that the appropriate Federal banking agencies, the Securities and Exchange Commission, and the Commodity Futures Trading Commission . . . determine [by rule].
Id. § 1851(h)(4). Reference to a “trading account” is intended to primarily cover short-term trades,
though federal regulators could expand that coverage. See id. § 1851(h)(6) (defining a trading account
as “any account used for acquiring or taking positions in the securities and instruments [described in the
definition of proprietary trading] principally for the purpose of selling in the near term (or otherwise
with the intent to resell in order to profit from short-term price movements), and any such other
accounts as the appropriate Federal banking agencies, the Securities and Exchange Commission, and
the Commodity Futures Trading Commission . . . determine [by rule]”).
2013] RING-FENCING 81
fund.”79 The regulatory implementation of the Volcker Rule, however, has
been significantly weakened by numerous exceptions and variances.80
A European Commission-appointed panel of experts, chaired by Bank
of Finland governor Erkki Liikanen, recently promulgated a report (the
Liikanen Report81) that has certain parallels to both the Volcker Rule and
the Vickers Report. Although the Liikanen Report does not refer to ring-
fencing, it recommends that banks separate certain risky activities from
deposit-taking.82 Subject to a materiality threshold, deposit-taking banks
could engage in proprietary trading and the taking of asset or derivative
positions in the process of market-making only through a separate “trading
entity.”83 Moreover, only that separate entity, and not a deposit-taking
bank, could extend credit to hedge funds, structured investment vehicles,
and private equity funds.84 The Liikanen Report therefore effectively
recommends ring-fencing to limit firms—in this case, deposit-taking
banks—from engaging in risky activities and making risky investments,
similar to the goals of the Volcker Rule and the Vickers Report.85
E. FUNCTIONAL DEFINITION
The foregoing discussion has shown that, functionally, ring-fencing
has at least four uses: to protect a firm from becoming subject to liabilities
and other risks associated with bankruptcy; to help ensure that a firm is
79. Id. § 1851(a)(1)(B). Notwithstanding these restrictions, trading is permitted “in connection
with underwriting or market-making, to the extent that either does not exceed near term demands of
clients, customers, or counterparties; on behalf of customers; or by an insurance business for the general
account of the insurance company.” Sweet & Christiansen, supra note 77, at 2.
80. Kimberly D. Krawiec, Don’t “Screw Joe the Plumber”: The Sausage-Making of Financial
Reform, 55 ARIZ. L. REV. 53, 68–70 (2013). The Volcker rule has faced significant criticism in the
United States. See, e.g., JAMES R. BARTH & APANARD PRABHA, MILKEN INST., BREAKING (BANKS) UP
IS HARD TO DO: NEW PERSPECTIVE ON ‘TOO BIG TO FAIL’ 24–26 (2013), available at
https://www.milkeninstitute.org/pdf/BreakingBanks.pdf (discussing some of the criticisms to the
Volcker Rule including the potential to “reduce liquidity and increase transaction costs,” and the
potential that the Volcker Rule targets the wrong firms because an analysis of the fifteen largest trading
losses since 1990 reveals that the largest losses were at nonbank financial firms).
81. HIGH-LEVEL EXPERT GROUP ON REFORMING THE STRUCTURE OF THE EU BANKING SECTOR,
FINAL REPORT (2012), available at http://ec.europa.eu/internal_market/bank/docs/high-
level_expert_group/liikanen-report/final_report_en.pdf [hereinafter EU BANK PANEL REPORT].
82. Id. at i.
83. Id. at 101.
84. Id.
85. As this Article was being finalized, France enacted legislation requiring limited ring-fencing
of deposit-taking banks. See Client Memorandum from Davis Polk & Wardwell LLP, France Ring-
Fences Proprietary Trading Activities (July 30, 2013),
http://www.davispolk.com/sites/default/files/07.30.13.Ring_Fences_1.pdf (summarizing the ring-
fencing aspects of France’s Banking Reform of July 27, 2013).
82 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
able to operate on a standalone basis even if its affiliated firms fail; to
protect a firm from being taken advantage of by affiliated firms, thereby
preserving the firm’s business and assets; and to limit a firm from engaging
in risky activities.86 In each case, law, including contracting, is used to
achieve the ring-fencing. Drawing on these uses, this Article will
tentatively define ring-fencing as legally deconstructing a firm—viewing a
“firm” broadly as a nexus-of-contracts87—to reallocate and reduce risk
more optimally,88 such as by protecting the firm’s assets and operations and
minimizing its internal and affiliate risks.
This definition still needs clarification because certain uses of ring-
fencing, such as ring-fencing used in securitization transactions and in
some covered bond transactions,89 are voluntarily undertaken by private
parties, whereas other uses of ring-fencing are required by government
86. Ring-fencing can be viewed as also having additional uses. One commentator suggests, for
example, that it has a fifth function: [M]aking the job of the regulator easier by simplifying market structure at the micro-level of the firm to the macro-level of the entire financial system. . . . [T]he simpler the structure of the [firm,] the less time it takes to investigate and collect strong evidence to support enforcement action. Furthermore, the less hesitant the regulator will be to pursue enforcement. In the past regulators hesitated out of fear of the systemic implications of the enforcement action.
Email from Andromachi Georgosouli, Lecturer, Queen Mary, Univ. of London, to author (July 4, 2013,
8:52 AM) (on file with author). Ring-fencing can also be used, in a variant of subsidiarization, see infra
text accompanying notes 195–97, to protect against cross-border risk by structuring a firm’s
international operations through separately capitalized subsidiaries. Lawrence Baxter, Size,
Subsidiarization and Stability, THEPARETOCOMMONS (Jan. 24, 2011),
http://www.theparetocommons.com/2011/01/size-subsidiarization-and-stability/. That use of ring-
fencing can be abused, however, if it is intended to allow, or has the effect of allowing, the foreign
subsidiaries to operate without adequate capital. Daniel K. Tarullo, Member, Bd. of Governors of the
Fed. Reserve Sys., Speech at the Yale School of Management Leaders Forum: Regulation of Foreign
Banking Organizations (Nov. 28, 2012), available at
http://www.federalreserve.gov/newsevents/speech/tarullo20121128a.pdf (recommending that foreign
banks operating in the United States through subsidiaries be required to establish a U.S.-based
intermediate holding company (IHC) to prevent those banks from avoiding U.S. consolidated capital
regulations).
87. According to the nexus-of-contracts theory of corporations, [T]he corporation [is] a bundle of market-driven actual and hypothetical bargains among shareholders, managers, and other firm participants, including outside third parties that deal with the firm. Neither corporations nor their shareholders are thought of as having external moral or social obligations independent of contract—the corporation because it is not a person, and the shareholders because they do not contract for broader responsibilities.
J. William Callison, Rationalizing Limited Liability and Veil Piercing, 58 BUS. LAW. 1063, 1065 (2003)
(footnote omitted). See generally Michael C. Jensen & William H. Meckling, Theory of the Firm:
Managerial Behavior, Agency Costs and Ownership Structure, 3 J. FIN. ECON. 305 (1976) (discussing
the nexus-of-contracts theory of corporations).
88. By “more optimally,” this Article means more socially optimally. Deconstructing a firm to
reallocate and reduce risk solely from the firm’s standpoint, regardless of externalities, is a form of
judgment proofing. See infra text accompanying notes 92–95.
89. See supra text accompanying notes 24–34.
2013] RING-FENCING 83
regulation.90 Although the term ring-fencing can broadly refer to all these
uses, this Article focuses on “regulatory” uses of ring-fencing—that is,
ring-fencing that is required by government regulation.91
That focus also helps to distinguish ring-fencing from “judgment
proofing.” The latter term refers to strategies taken by firms to externalize
costs by separating their ownership of assets from the liabilities associated
with operating those assets.92 To that extent, both ring-fencing and
judgment proofing involve a firm’s deconstruction. In contrast, however, to
regulatory uses of ring-fencing93 (and also in contrast to many ring-fencing
transactions that are voluntarily undertaken by private parties94), the goal of
judgment proofing is to impose externalities on a firm’s creditors,
preventing them from enforcing their claims against assets that otherwise
should be available for payment.95
III. NORMATIVE ANALYSIS
Why should ring-fencing be used as a regulatory tool? Being a form of
financial regulation,96 ring-fencing is a subset of economic regulation.97
Economic regulation has two fundamental normative goals. Ordinarily,
economic regulation is intended to help correct market failures98 within the
financial system.99 Absent such failures, financial markets should operate
90. Most of the examples used in Part II, including utility ring-fencing and the ring-fencing of
banks under the Vickers Report and the Glass-Steagall Act, involve ring-fencing that is required by
government regulation. See supra Part II.
91. Regulatory uses of ring-fencing can include ring-fencing that is required by government
regulation but implemented contractually. See supra text accompanying note 36 (observing that the
regulatory ring-fencing of utility companies is implemented through a combination of contract and
legislation).
92. Steven L. Schwarcz, The Inherent Irrationality of Judgment Proofing, 52 STAN. L. REV. 1, 2
(1999) (citing Lynn M. LoPucki, The Death of Liability, 106 YALE L.J. 1, 4 (1996)).
93. This Article assumes that regulatory uses of ring-fencing will not have the goal of imposing
externalities.
94. Schwarcz, supra note 92, at 12–17 (distinguishing legitimate securitization transactions from
judgment proofing).
95. Id. at 4–10.
96. See supra text accompanying notes 11–20 (examining ring-fencing as a financial regulatory
concept).
97. Cf. VICKERS REPORT, supra note 2, at 35–77 (discussing ring-fencing as a type of economic
regulation).
98. PAUL A. SAMUELSON & WILLIAM D. NORDHAUS, ECONOMICS 756 (15th ed. 1995) (defining
market failure as “[a]n imperfection in a price system that prevents an efficient allocation of
resources”). See also JOHN BLACK, NIGAR HASHIMZADE & GARETH MYLES, A DICTIONARY OF
ECONOMICS 283 (3d ed. 2009) (defining market failure).
99. E.g., DAVID GOWLAND, THE REGULATION OF FINANCIAL MARKETS IN THE 1990S, at 21
(1990). Welfare economists argue that regulation should also include the goal of maximizing social
welfare. See, e.g., Charles Wolf, Jr., A Theory of Nonmarket Failure: Framework for Implementation
84 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
efficiently without any regulation.100 Economic regulation can also help to
protect against “risks to the financial system itself.”101 These types of risks
are referred to as “systemic,” and they “transcend[] economic efficiency
per se.”102
This Article next examines ring-fencing in the context of market
failures and efficiency.103 Thereafter, it examines ring-fencing as a possible
protection against systemic risk.104
A. RING-FENCING TO CORRECT MARKET FAILURES
The market failures potentially relevant to economic regulation are
(1) monopolies and other forms of noncompetitive markets, (2) the public-
goods problem, (3) information failure, (4) agency failure, and
(5) externalities.105 The analysis below examines ring-fencing in light of
these market failures, subject to a clarification.
It is confusing to regard “externalities” as a separate category of
market failure. One source of confusion is that externalities are
consequences, not causes, of market failure.106 Their only link to causation
is to signal that a market failure has occurred.107 Another source of
confusion is that externalities cannot even be linked to a distinct category
of market failure because “all types of market failures can result in
externalities.”108 To avoid these confusions, this Article will refer to the
final category of market failure not as “externalities” per se but, consistent
with recent scholarship,109 as “responsibility failure”—meaning a firm’s
Analysis, 22 J.L. & ECON. 107, 110–11 (1979). For a general discussion of the justifications for
economic regulation, see STEPHEN BREYER, REGULATION AND ITS REFORM 15–35 (1982) (explaining
that “the justification for intervention arises out of an alleged inability of the marketplace to deal with
particular structural problems”).
100. Cf. IVAN PNG & DALE LEHMAN, MANAGERIAL ECONOMICS 414 (3d ed. 2007) (observing
that government regulation enhances social welfare by correcting market failures).
101. Steven L. Schwarcz, Systemic Risk, 97 GEO. L.J. 193, 207 (2008).
102. Id.
103. See infra Part III.A.
104. See infra Part III.B.
105. See Richard O. Zerbe Jr. & Howard E. McCurdy, The Failure of Market Failure, 18 J. POL’Y
ANALYSIS & MGMT. 558, 561 (1999) (discussing negative externalities, monopolies, and information
asymmetries); Steven L. Schwarcz, Regulating Shadows: Financial Regulations and Responsibility
Failure, 70 WASH. & LEE L. REV. 1781, 1785–86 (discussing information failure, agency failure, and
externalities).
106. Schwarcz, supra note 105, at 1800–01.
107. Id.
108. Id. at 1801.
109. See id. at 1799–804 (arguing that “responsibility failure” should be a separate category of
market failure, in lieu of “externalities”).
2013] RING-FENCING 85
ability to externalize all or a portion of the costs of taking an action.
1. Monopolies and Other Forms of Noncompetitive Markets
A monopoly is a market condition where only one supplier or
producer has exclusive control over the commercial market within a given
region.110 The traditional economic rationale for regulation of a monopolist
is that an unregulated monopolist will restrain production to retain higher
prices.111 The result is unfair pricing and undersupply.112 Additional bases
for regulation of a monopolist include price discrimination, income transfer
from users of the service to investors, fairness (more than just price
discrimination), and power (specifically fear of concentration of power).113
Other forms of noncompetitive markets include oligopolies. An
oligopoly is a market controlled by a small group of firms.114 An oligopoly
can occur when the pricing and output policies of firms are
interdependent.115 Firms therefore are able to collude to maximize joint
profits through quantity or price setting.116 The rationale for regulating an
oligopoly is therefore similar to monopoly regulation: to avoid undersupply
and unfair pricing.
Financial firms are not usually subject to this category of market
failure, however. The market for financial firms, even insofar as it pertains
to regulated banking activities, is in fact competitive.117 Furthermore, even
though ring-fencing is otherwise strongly associated with this category of
market failure, that association is coincidental. Utility companies—which
historically are the firms most subject to ring-fencing118—are monopolies.
Nonetheless, ring-fencing’s application to utilities is relevant not to
110. See BLACK’S LAW DICTIONARY 1098 (9th ed. 2009) (defining “monopoly” as “1. Control or
advantage obtained by one supplier or producer over the commercial market within a given
region. . . . 2. The market condition existing when only one economic entity produces a particular
product or provides a particular service”).
111. BREYER, supra note 99, at 15–16 (discussing the traditional economic rationale for regulation
of a monopoly).
112. Id.
113. Id. at 17–20.
114. Glossary of Statistical Terms: Oligopoly, OECD, http://stats.oecd.org/glossary/
detail.asp?ID=3270 (last updated Mar. 10, 2003).
115. Id.
116. Id.
117. See, e.g., Harry Terris, The Nation’s Most, and Least, Competitive Banking Markets, AM.
BANKER (May 9, 2011), http://www.americanbanker.com/issues/176_89/competitive-banking-markets-
1037222-1.html (discussing competition in America’s banking markets and determining that while
some banking markets may be less competitive than others, the banking market as a whole is a
competitive market).
118. See supra Part II.
86 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
correcting unfair pricing but to addressing the risks associated with a
holding company structure.119 Additionally, as discussed below,120 ring-
fencing addresses the other market failures that afflict financial firms.
2. The Public-Goods Problem
The public-goods problem is a collective action supply problem,
resulting in either oversupply or undersupply. The typical solution to the
public-goods problem is government intervention—either providing the
goods directly (and taxing their cost) or requiring the private sector to
provide the goods. In each case, such government action is defined as
“public provision” of the goods.
The public-goods problem can arise when “goods,” in the broadest
sense of the word, have two characteristics: nonrivalry in consumption (use
of the goods by one person or group does not distract from their use for
other persons or groups) and nonexcludability (the benefits of the goods
cannot be reserved for use by one person or group).121
There are two forms of the public-goods problem: the free rider
problem and the prisoner’s dilemma.122 The free rider problem is the
situation in which persons or groups lack incentive to, or are incentivized
not to, contribute personal resources to common endeavors, free riding
instead off others’ efforts.123 The prisoner’s dilemma problem is the game
theory problem where persons or groups lacking the ability to communicate
make suboptimal decisions.124
119. See supra text accompanying notes 37–41. Utility pricing is typically set by the utility’s
applicable public service commission. See Douglas N. Jones, Agency Transformation and State Public
Utility Commissions, 14 UTIL. POL’Y 8, 9–11 (2006) (explaining that state public utility commissions
were created by legislatures to respond to excessive prices by utility companies, and also discussing
how state public utility commissions oversight of prices was relaxed in the 1990s).
120. See infra Part III.A.2–5 (discussing the use of ring-fencing to address the public-goods
problem, information failure, agency failure, and responsibility failure).
121. Inge Kaul, Isabelle Grunberg & Marc A. Stern, Defining Public Goods, in GLOBAL PUBLIC
GOODS: INTERNATIONAL COOPERATION IN THE 21ST CENTURY 2, 2–3 (Inge Kaul et al. eds., 1999). An
example of goods that meet both of these characteristics is a traffic light. Id. at 4. The use of the traffic
light by a pedestrian to safely cross the street does not distract from the light’s utility to other
pedestrians or drivers (nonrivalry in consumption). Id. And the benefit of the light cannot be reserved
for use by only one person or group (nonexcludability). Id.
122. Id. at 6.
123. Id. at 6–7.
124. Id. at 7–8. See also BLACK’S LAW DICTIONARY, supra note 110, at 1314–15 (defining
prisoner’s dilemma as “[a] logic problem—often used by law-and-economics scholars to illustrate the
effect of cooperative behavior—involving two prisoners who are being separately questioned about
their participation in a crime: (1) if both confess, they will each receive a 5-year sentence; (2) if neither
confesses, they will each receive a 3-year sentence; and (3) if one confesses but the other does not, the
confessing prisoner will receive a 1-year sentence while the silent prisoner will receive a 10-year
2013] RING-FENCING 87
The Public-Goods Problem and Ring-Fencing Financial Firms. How
does the public-goods problem apply to banks and other financial firms,
and can ring-fencing help to solve the problem? The most potentially
relevant “goods” are banking functions deemed important to society that
are normally provided by private-sector banks.125 These functions include
safeguarding “deposits, operating secure payments systems, efficiently
channelling savings to productive investments [making loans], and
managing financial risk.”126
Of these functions, only safeguarding deposits appears to suffer from a
public-goods problem.127 Safeguarding deposits is nonrivalrous because a
sentence”).
Returning to the traffic light example, suppose Resident A lives next to the intersection
where the traffic light would be located and therefore would benefit significantly more than Resident B,
who lives further away. If all residents on the street are asked to pay the same amount, Resident B may
rationally decide to refuse (usually referred to in the literature as “defection”). Defection results from
imperfect communication because the residents are unable to communicate to choose the outcome that
is best for all of them. This defection would result in undersupply if it deprives the residents of
sufficient funds to put up the light. Again, government provision of traffic lights would overcome this
problem by providing traffic lights and taxing persons to pay for the lights.
Sometimes, the government intervenes to solve the public-goods problem other than through
public provision of goods. A common approach, exemplified by the patent system, is to impose laws
that take certain critical goods out of the public-goods realm. Consider, for example, microcomputer
software. Randall G. Holcombe, A Theory of the Theory of Public Goods, 10 REV. AUSTRIAN ECON. 1,
7 (1997). Microcomputers, sometimes called personal computers, are computers designed for use by
individuals. Such software is nonrivalrous because additional users could utilize the software without
impairing its use by existing users. Id. Absent the patent system, the software is also nonexcludable
because it is costly to prevent such additional use. Id. Therefore, parties could free ride off the work of
software producers, depriving those producers of optimal compensation, thereby resulting in
undersupply of software. Id. at 8. Government normally solves this public-goods problem by enabling
software producers to patent their innovations, thus making the software excludable unless additional
users are willing to pay. Id.
125. The banking function of acting as a “lender of last resort” does not normally raise a public-
goods problem because that function is performed by government central banks.
126. VICKERS REPORT, supra note 2, at 7. See also BIAGIO BOSSONE, THE WORLD BANK, WHAT
MAKES BANKS SPECIAL? A STUDY ON BANKING, FINANCE AND ECONOMIC DEVELOPMENT 5–23
(1999) (discussing banks’ special role in the economy, including running the economy’s payments
system, portfolio and risk management, supplying credit, and providing liquidity).
127. Operating secure payments systems does not suffer from a public-goods problem because it
is not nonrivalrous: there is a limited amount of capital that can be used to operate the secure payments
system. See BD. OF GOVERNORS OF THE FED. RESERVE SYS., FEDERAL RESERVE POLICY ON PAYMENT
SYSTEM RISK (2011), http://www.federalreserve.gov/paymentsystems/files/psr_policy.pdf (explaining
Federal Reserve policy to limit payment system risk, including capital limits). It also is not
nonexcludable because individual financial firms can limit the benefits to only their customers. See,
e.g., Transferring Funds FAQs, BANK AM., https://www.bankofamerica.com/onlinebanking/electronic-
funds-transfer-faqs.go (last visited Nov. 2, 2013) (explaining that its electronic funds transfer options
are available only to customers of Bank of America and in-network accounts). Making loans does not
suffer from a public-goods problem because it is not nonrivalrous. A bank that makes a loan to
customer A will have less capital left to make a loan to customer B. See Basel Regulatory Capital
88 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
bank’s safeguarding such deposits for one person or group does not distract
from the bank’s ability to safeguard deposits for other persons or groups.
Those other persons or groups could also safeguard their deposits with
competing banks. Additionally, safeguarding deposits is nonexcludable.
The benefits of safeguarding deposits cannot be reserved for use by one
person or group because the market for banking, including taking and
safeguarding deposits, is competitive.128
Safeguarding deposits therefore could be subject to a public-goods
problem, and indeed it sometimes faces a prisoner’s dilemma problem,
causing suboptimal safeguarding.129 Although banks can and do
communicate and play a meaningful role in disciplining other banks,
especially regarding risk management,130 interbank discipline alone cannot
optimize the safeguarding of deposits. For example, banks cannot perfectly
monitor other banks about which they have imperfect information.131 There
therefore is a need for government intervention to improve the
safeguarding of deposits. In the United States, the government does this
through Federal Deposit Insurance Corporation (“FDIC”) deposit
insurance.132
The government could further safeguard deposits through ring-
fencing. This could occur in various ways, such as by legally isolating
deposit-taking banks from liabilities associated with riskier banking
activities and from insolvency risks, or by giving depositor claims legal
priority over the claims of other bank creditors.133
Framework, BOARD GOVERNORS FED. RES. SYS., http://www.federalreserve.gov/bankinforeg/basel/
default.htm (last updated Oct. 25, 2013) (discussing the capital that banks are required to hold to absorb
losses and thus cannot be used to make loans). Making loans is also not nonexcludable. Banks exclude
customers, for example, through credit checks. See, e.g., Mortgage Prequalification Request, CHASE,
https://apply.chase.com/Mortgage/gettingstarted.aspx (last visited Nov. 1, 2013) (disclaiming that “All
loans subject to credit and property approval”). Managing financial risk likewise does not suffer from a
public-goods problem. It is nonrivalrous because the benefits of managing customer A’s risk does not
distract from the benefits of managing customer B’s risk. It is not nonexcludable because it is a service
limited only to bank customers.
128. See supra note 117 and accompanying text.
129. Recall that this problem can occur where persons or groups lacking the ability to
communicate make suboptimal decisions. See supra note 124 and accompanying text.
130. See Kathryn Judge, Interbank Discipline, 60 UCLA L. REV. 1262, 1286–92 (2013)
(discussing how banks have taken on an expanding role in the discipline of other banks). Banks can
discipline other banks, for example, by limiting economic exposure to those banks. Id. at 1289.
131. Id. at 1299.
132. See Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111-203,
§ 335, 124 Stat. 1376, 1540 (2010) (permanently increasing deposit insurance to $250,000).
133. Deposit accounts are, technically, claims by depositors against the bank. See Bank Liabilities,
AMOSWEB, http://www.amosweb.com/cgi-bin/awb_nav.pl?s=wpd&c=dsp&k=bank%20liabilities (last
visited Nov. 1, 2013) (discussing customer deposits as the most important category of bank liability).
2013] RING-FENCING 89
3. Information Failure
Information failure, a type of market failure that results from
inadequate information, plagues financial firms.134 One form of
information failure is asymmetric information, which occurs when a party
in a transaction has an information advantage over another party.135 That
can result in harm if the party with superior information uses the
asymmetry to take advantage of the other party.136 For example, issuers of
securities have more information about the securities they issue than
investors in those securities.137 Without disclosing this information to
investors, an issuer of securities could sell the securities for more than they
are worth. To resolve this information failure and protect investors,
securities law requires mandatory disclosures by issuers.138
Complexity exacerbates the disclosure problems of asymmetric
information.139 Financial markets and transactions have become
increasingly complex.140 In some cases, the complexity undermines the
ability of disclosure to achieve meaningful transparency.141 For example,
during the recent financial crisis most of the risks on complex mortgage-
backed securities were disclosed.142 Despite these disclosures, “investors—
including even the largest, most sophisticated firms—bought these
securities without fully understanding them.”143
One might ask why sophisticated firms cannot hire experts to help
them understand complex financial products. Part of the reason is that, as
complexity increases, a larger amount of information must be incorporated
into risk analysis to “value the investment with a degree of certainty.”144
134. See BREYER, supra note 99, at 26–28 (discussing inadequate information and the rationales
for regulation). There is some overlap among market-failure categories. The prisoner’s dilemma
problem, for example, results in part from inadequate information in the form of inadequate
communication. See supra note 124 and accompanying text.
135. See Asymmetric Information, INVESTOPEDIA, http://www.investopedia.com/terms/a/
asymmetricinformation.asp (last visited Nov. 1, 2013).
136. Id.
137. Cf. Frank H. Easterbrook & Daniel R. Fischel, Mandatory Disclosure and the Protection of
Investors, 70 VA. L. REV. 669, 714–15 (1984) (discussing information asymmetries between issuers and
investors and arguing that disclosure is the principal justification for the federal securities laws).
138. Id.
139. See Steven L. Schwarcz, Controlling Financial Chaos: The Power and Limits of Law, 2012
WIS. L. REV. 815, 818–21.
140. Id. at 818.
141. Id. at 818–19.
142. Id.
143. Id. at 819.
144. Steven L. Schwarcz, Regulating Complexity in Financial Markets, 87 WASH. U. L. REV. 211,
221 (2009).
90 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
This type of analysis requires additional resources of time and staff, which
may “outweigh[] the uncertain gain.”145
A solution to the problems posed by complexity includes
standardization of investments.146 Standardization, however, can backfire
by “stifl[ing] innovation” and preventing parties from “craft[ing] financial
products [that are] tailored to [their] particular needs and risk
preferences.”147
Another form of information failure is the problem of “bounded
rationality.”148 People are not wholly rational actors.149 We have difficulty,
for example, appreciating unlikely events that, if they occur, could have
devastating consequences.150 This bounded rationality causes information
failure: people misinterpreting, overrelying, or underrelying on
information.151 For example, due to familiarity with collateral, members of
the financial community “underestimate[d] the likelihood and potential
consequences of a drop in housing prices.”152 This drop in collateral value
turned what was thought to be overcollateralized mortgage-backed
securities into undersecured securities.153
Information Failure and Ring-Fencing Financial Firms. Financial
firms suffer from both forms of information failure: asymmetric
information and bounded rationality. They suffer from asymmetric
information when issuers of securities have more information about the
underlying investment than investors in the securities.154 Although
securities law disclosure requirements seek to resolve this asymmetric
information problem, the asymmetry can be exacerbated by complexity.155
Ring-fencing can help to address this type of information failure, such as
145. Id. at 221–22.
146. Schwarcz, supra note 139, at 820.
147. Id.
148. See Schwarcz, supra note 105, at 1791–92.
149. Schwarcz, supra note 139, at 821–22, 825.
150. Iman Anabtawi & Steven L. Schwarcz, Regulating Systemic Risk: Towards an Analytical
Framework, 86 NOTRE DAME L. REV. 1349, 1366–68 (2011).
151. Cf. Schwarcz, supra note 139, at 821 (“Even in financial markets, humans have bounded
rationality—a type of information failure . . . .”).
152. See id. at 822.
153. Id. The Dodd-Frank Act attempts to solve this bounded rationality problem by improving the
quality of rating-agency ratings. Id.
154. See supra text accompanying notes 134–38.
155. See, e.g., Steven L. Schwarcz, Disclosure’s Failure in the Subprime Mortgage Crisis, 2008
UTAH. L. REV. 1109, 1113 (discussing how complexity can undermine the effectiveness of disclosure
requirements).
2013] RING-FENCING 91
by simplifying the investments that certain financial firms can make.156
Financial firms also experience information failure in the form of
bounded rationality. Bounded rationality can impact financial firms in the
form of bank runs.157 In a bank run, some depositors panic, converging on
the bank in a “grab race” to withdraw their monies first. Because banks
keep only a small fraction of their deposits on hand as cash reserves, other
depositors may have to join the run in order to avoid losing the grab
race.158 If there is insufficient cash to pay all withdrawal demands, the bank
will default.159 In effect, the bounded rationality of individuals fearing a
bank run can create a self-fulfilling prophecy. Ring-fencing financial firms
can address this problem of bounded rationality. Ring-fencing the essential
banking functions of financial firms, specifically deposit-taking, insulates
deposits from the legal liabilities and insolvency risks caused by financial
firms’ other riskier activities. Consequently, deposits will be more stable
and less prone to suffering from instability caused by activities of the
financial firm. This added stability should make depositors less likely to
panic, thereby reducing the risk of bank runs.
4. Agency Failure
Because it impacts the management of financial firms, agency failure
is a type of market failure that is relevant to economic regulation. The
following will analyze agency failure in that context, examining whether
ring-fencing can help to correct the failure.
In general, agency failure can exist whenever there is a conflict of
interest between principals and their agents.160 The well-known principal-
agent conflict in this Article’s context is between the owners, typically
156. This is in part the intention of the Volcker Rule. See Letter from Paul A. Volcker to Timothy
Geithner, Chairman, Fin. Stability Oversight Council (Oct. 29, 2010) (“The plain intent of Section 619
of the Dodd-Frank Act [12 U.S.C. § 1851(a)(1), the Volcker Rule] is to restrict certain high risk,
proprietary trading activities by banks and bank holding companies, institutions that receive
government protection and support.”).
157. Cf. Douglas W. Diamond & Philip H. Dybvig, Bank Runs, Deposit Insurance, and Liquidity,
91 J. POL. ECON. 401, 404 (1983) (using the Diamond-Dybvig model to explain bank runs as a form of
undesirable equilibrium triggered by expectations based on incomplete information, in which depositors
(sometimes irrationally) expect the bank to fail, thereby causing its failure). Information failures
arguably are only part of the cause of bank runs, as will be addressed in Part III.A.5.
158. See, e.g., Jonathan R. Macey & Geoffrey P. Miller, Bank Failures, Risk Monitoring, and the
Market for Bank Control, 88 COLUM. L. REV. 1153, 1156 (1988) (linking bank runs and depositor
collective action problems).
159. R.W. HAFER, THE FEDERAL RESERVE SYSTEM: AN ENCYCLOPEDIA 145 (2005) (observing
that a bank’s cash reserves are often less than five percent of its deposits).
160. See, e.g., Lucian A. Bebchuk & Jesse M. Fried, Pay Without Performance: Overview of the
Issues, 20 ACAD. MGMT. PERSP. 5, 7–8 (2006) (discussing the classic principal-agent conflict).
92 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
shareholders, and managers of a firm.161 However, an additional, and
conceivably more important, agency problem can arise intrafirm—between
middle managers and the senior managers to whom they report.162 Middle
managers are typically paid under short-term compensation schemes, in
which they are entitled to keep their compensation for work performed in
any individual year even if that work later results in significant losses for
the firm.163 This misaligns their interests with the long-term interests of the
firm.164 As a result, even firms with reputations for highly sophisticated
risk management, such as JPMorgan, have proven susceptible to failures
leading to significant losses.165
A number of solutions seek to address the problems of agency failure.
These include regulations that prevent bank managers from taking risks
that benefit them more than their banks.166 Securities law and corporation
law create fiduciary duties of managers to shareholders.167 Commentators
have also been proposing a more long-term realignment of managerial
compensation with interests of the firm.168
Agency Failure and Ring-Fencing Financial Firms. There does not
161. Id.
162. Steven L. Schwarcz, Conflicts and Financial Collapse: The Problem of Secondary-
Management Agency Costs, 26 YALE J. ON REG. 457, 457–58 (2009).
163. See, e.g., Steven M. Davidoff, After $2 Billion Loss, Will JPMorgan Move to Claw Back
Pay?, N.Y. TIMES DEALBOOK (May 14, 2012), http://dealbook.nytimes.com/2012/05/14/after-2-billion-
trading-loss-will-jpmorgan-claw-back-pay/ (discussing how, even after the recent institution of a
clawback policy, traders asked to leave due to large losses may be able to keep previous compensation
in the millions of dollars).
164. Schwarcz, supra note 162, at 462.
165. Consider, for example, JPMorgan’s recent $5.8 billion trading loss. Christine Harper,
JPMorgan Loss Proves System Too Complex, China’s Gao Says, BLOOMBERG BUSINESSWEEK (Oct. 05,
2012), http://www.businessweek.com/news/2012-10-05/jpmorgan-loss-proves-system-too-complex-
china-s-gao-says. The loss was due to allegedly insufficient oversight over a trader, tarnishing
JPMorgan’s reputation for being a “strong risk manager.” Tom Braithwaite & Ajay Makan, JPMorgan
Revamps Board Risk Committee, FIN. TIMES (May 25, 2012), http://www.ft.com/intl/cms/s/0/6702908e-
a675-11e1-9453-00144feabdc0.html#axzz2dIeGuK8C. See also Gregory Zuckerman & Dan
Fitzpatrick, ‘Whale’ Swam in Choppy Waters, WALL ST. J. (Jun. 19, 2012),
http://online.wsj.com/news/articles/SB10001424052702303379204577474842039937860 (stating that
the $2 billion loss “tarred the reputation of [JPMorgan] Chief Executive James Dimon as Wall Street’s
savviest risk manager”).
166. See, e.g., Arthur E. Wilmarth, Jr., The Transformation of the U.S. Financial Services
Industry, 1975–2000: Competition, Consolidation, and Increased Risks, 2002 U. ILL. L. REV. 215, 264
(comparing internal loan regulations in large and small banks).
167. Schwarcz, supra note 105, at 1790–91.
168. Schwarcz, supra note 162, at 465–67. See also Bebchuk & Fried, supra note 160, at 18–20
(providing proposals for making executive pay, and its relationship to performance, more transparent);
Schwarcz, supra note 105, at 1790 (discussing improvements in corporate governance as tools to reduce
conflicts of interest).
2013] RING-FENCING 93
appear to be a significant role for ring-fencing in helping to correct agency
failure. Ring-fencing does not purport to address, at least directly, questions
of managerial compensation or conflicts of interest. Ring-fencing could be
used indirectly, though, to address those questions; for example, by limiting
the ability of managers of a financial firm to make risky investments,169
those managers could be limited from booking investments that pay them
bonuses but have long-term risks to the firm.170
5. Responsibility Failure
Recall that this category of market failure references a firm’s ability to
externalize all or a portion of the costs of taking an action.171 For example,
because the managers of most firms have obligations under law solely to
the firms’ shareholders, a firm that engages in a risky project in order to
increase shareholder profit opportunities may well be acting responsibly as
defined, indeed mandated, by law—even if the effect is to externalize
costs.172 The ability of a firm to so externalize costs is a market failure.173
The merit of the term “responsibility failure” is that it shifts focus onto
the party who should be fundamentally responsible for internalizing the
externality. Focusing on externalities, one may well conclude that the firm
itself in the preceding example should be considered solely responsible for
causing the externalities. Focusing on responsibility failure, in contrast,
would help shift attention back to the fundamental cause of the
externalities: in this case, the government’s failure to impose laws that limit
the ability of firms to externalize those costs.174 “This sharpened focus on
causation is important because the traditional paradigm of market failure
assumes away government action (or inaction) as a cause of failure.”175
Of the possible ways to address responsibility failure, the most direct
would be to try to require firms to internalize their externalities. There is
currently a debate, for example, whether the government should mandate
that financial firms, or at least financial firms that have the potential to
169. Cf. supra text accompanying note 74 (discussing ring-fencing to prevent risky investments
under the Volcker Rule).
170. Cf. supra note 163 and accompanying text (discussing short-term compensation that allows
managers to keep their compensation for work performed in any individual year even if that work later
results in significant losses for the firm).
171. See supra text accompanying note 109.
172. Schwarcz, supra note 105, at 1803.
173. See id. at 1816–17.
174. Cf. Zerbe & McCurdy, supra note 105, at 571 (observing that certain “markets are inefficient
not because of any inherent ‘failures,’ but because the government has neglected to provide the
appropriate institutional framework”).
175. Schwarcz, supra note 105, at 1802–03.
94 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
generate large externalities (such as systemically important financial
institutions (“SIFIs”)), contribute to a fund that would help to offset the
externalities.176 Although this should work in principle, it may be difficult
to price risk outside of actual markets, making the fund difficult to
implement.177 As explained below, ring-fencing could help to address the
problem of responsibility failure.
Responsibility Failure and Ring-Fencing Financial Firms. The
problem of responsibility failure could be addressed remedially, such as by
requiring financial firms to try to internalize any externalized costs, as
discussed above.178 The problem could also be addressed more directly,
176. See Schwarcz, supra note 139, at 830 (discussing this debate). A variant on that approach, in
line with the Pigouvian approach to externalities, would be to impose taxes to internalize the social
costs of activities. See Jules L. Coleman, Efficiency, Exchange, and Auction: Philosophic Aspects of the
Economic Approach to Law, 68 CALIF. L. REV. 221, 232 (1980) (arguing that externalities should be
controlled by taxes). Such a tax might seek to eliminate wasteful short-term currency speculation and
reduce market volatility by imposing a tax on individual financial trades. See 2 ENCYCLOPEDIA OF THE
WORLD ECONOMY 1093–94 (Kenneth A. Reinert et al. eds., Princeton Univ. Press 2009); Steven M.
Davidoff, In Wall St. Tax, a Simple Idea but Unintended Consequences, N.Y. TIMES DEALBOOK (Feb.
26, 2013), http://dealbook.nytimes.com/2013/02/26/in-wall-street-tax-a-simple-idea-with-unintended-
consequences/. A financial transactions tax has been the subject of heated debate. Compare Davidoff,
supra (articulating arguments against such a tax), with Thomas I. Palley, The Economic Case for the
Tobin Tax, in DEBATING THE TOBIN TAX: NEW RULES FOR GLOBAL FINANCE 5 (James Weaver et al.
eds., 2003) (articulating arguments for such a tax). Eleven eurozone countries—Germany, France, Italy,
Spain, Austria, Portugal, Belgium, Estonia, Greece, Slovakia, and Slovenia—are in the process of
implementing this type of tax. John O’Donnell & Robin Emmott, EU States Get Blessing for Financial
Trading Tax, REUTERS (Jan. 22, 2013), http://www.reuters.com/article/2013/01/22/us-eu-
transactionstax-idUSBRE90K0WX20130122. They also are pressuring the United States to adopt such
a tax. See, e.g., Carey L. Biron, Europeans Urge U.S. Action on Financial Transaction Tax, INTER
PRESS SERVICE (Feb. 25, 2013), http://www.ipsnews.net/2013/02/europeans-urge-u-s-action-on-
financial-transaction-tax/.
177. See Schwarcz, supra note 139, at 830 (discussing the potential obstacles to the creation of a
systemic risk fund). Another way that regulators could attempt to address responsibility failure is by
micromanaging firms, such as mandating leverage, liquidity, and investment requirements. The Dodd-
Frank Act requires banks and other systemically important financial firms to adhere to a range of capital
and similar requirements. Id. at 834. Although leverage requirements have the goal of enabling a firm to
withstand economic shocks, there is no optimal across-the-board amount of leverage. Id. Additionally,
the inability of the Basel capital requirements to prevent bank failures during the global financial crisis
raises doubt that the Dodd-Frank capital requirements will be any more successful. Id. Cf. Arthur E.
Wilmarth, Jr., The Dodd-Frank Act: A Flawed and Inadequate Response to the Too-Big-to-Fail
Problem, 89 OR. L. REV. 951, 1009–15 (2011) (discussing the problems with capital-based regulation,
including that capital ratios are “lagging indicators” and that firms have demonstrated their ability to
weaken the effectiveness of capital requirements by engaging in “regulatory capital arbitrage” (internal
quotation marks omitted)).
178. See supra notes 176–77 and accompanying text. Another direct approach would be simply to
make it illegal for a firm to externalize its costs, but it is difficult to conceive how that approach could
be made workable. Cf. EMILIOS AVGOULEAS, GOVERNANCE OF GLOBAL FINANCIAL MARKETS: THE
LAW, THE ECONOMICS, THE POLITICS 96–106 (Eilis Ferran et al. eds., 2012) (noting that failing to
internalize certain externalities is a market failure); Henry N. Butler & Jonathan R. Macey, Externalities
2013] RING-FENCING 95
however, by limiting the firm’s ability to externalize costs in the first place.
Ring-fencing could help implement that latter approach, such as by limiting
a financial firm’s risky activities and investments.179 For example, the
Volcker Rule is directed at limiting the ability of banks to make risky
investments.180 Avoiding those investments would help to deter bank
failures, thereby reducing the risk that such failures would lead to a
systemic collapse of the banking system.181
B. RING-FENCING TO PROTECT AGAINST SYSTEMIC RISK
Part III.A above has shown that financial firms are subject to a number
of market failures, and that ring-fencing can be used to help correct some
of those failures. This Part III.B, in contrast, examines ring-fencing as a
protection against systemic risk. Ring-fencing can help in two ways to
protect against systemic risk: by minimizing panics and by creating
modularity.
1. Minimizing Panics
Panics are a common trigger of systemic risk.182 Ring-fencing
therefore could reduce systemic risk if it could minimize panics. In today’s
disintermediated financial system—in which bank intermediation is no
longer needed to source funds from capital markets to firms that use the
funds to operate in (and thus contribute to) the real economy183—market
failures can easily be amplified to cause panics.184 By correcting market
and the Matching Principle: The Case for Reallocating Environmental Regulatory Authority, 14 YALE
L. & POL’Y REV. 23, 29–30 (1996) (noting in the context of pollution control—a negative externality—
“it is important to recognize that the combination of small externalities and nontrivial costs of
government intervention suggests that many externalities cannot be internalized”); Brett M.
Frischmann, An Economic Theory of Infrastructure and Commons Management, 89 MINN. L. REV. 917,
967 (2005) (“Neither the law nor economic efficiency require complete internalization; external
benefits are a ubiquitous boon for society.”).
179. See supra Part II.D (discussing that function of ring-fencing).
180. See supra notes 78–79 and accompanying text.
181. Chris Mundy, The Nature of Risk: The Nature of Systemic Risk—Trying to Achieve a
Definition, 12 BALANCE SHEET, no. 5, 2004, at 29, 29.
182. Schwarcz, supra note 101, at 214–18.
183. See Steven L. Schwarcz, Regulating Shadow Banking: Inaugural Address for the Inaugural
Symposium of the Review of Banking & Financial Law, 31 REV. BANKING & FIN. L. 619, 624–25
(2012). The term “disintermediation” is, to some extent, a misnomer because there still may be nonbank
intermediaries between financial markets and users of funds. Those nonbank intermediaries include
special-purpose entities and other entities that operate without access to central bank liquidity or public
sector credit guarantees, including finance companies, hedge funds, money-market mutual funds,
securities lenders, and investment banks. Id. at 621.
184. See, e.g., Dan Awrey, Complexity, Innovation, and the Regulation of Modern Financial
Markets, 2 HARV. BUS. L. REV. 235, 267–77 (2012) (analyzing how financial innovation can increase
96 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
failures, ring-fencing thus could help to minimize panics.185
Recall that ring-fencing can help to correct market failures by
reducing information asymmetry,186 safeguarding deposit-taking functions
of banks,187 and limiting the ability of financial firms to engage in risky
behavior or make risky investments.188 Correcting these failures not only
would increase market efficiency, it also would prevent the failures from
being amplified into panics, thereby protecting against systemic risk.
2. Creating Modularity
Ring-fencing can also help to protect against systemic risk by creating
modularity. The financial system is highly complex,189 and failures are
almost inevitable in complex systems.190 Chaos theory—more technically
known as the theory of complex adaptive systems—posits, however, that
complex systems can be made more successful by limiting the
consequences of a failure.191 This can be accomplished by decoupling the
system through “modularity,” helping to reduce the chance that a failure in
one part of the system will systemically trigger a failure in another part.192
Ring-fencing could insert modularity into the financial system by
using some or all of the tools discussed—including bankruptcy remoteness,
ability to operate on a standalone basis, protection against affiliates,
limitations on risky activities and investments, and protection against cross-
border risks193—to protect certain systemically important financial firms.194
That would help to ensure that failures of those firms’ affiliates or
counterparties would not necessarily cause the ring-fenced firms to fail.
The Vickers Report195 implicitly refers to this as the use of
complexity in financial markets and result in pervasive information asymmetries and expertise
asymmetries). In an interconnected financial market, financial shocks can be transmitted faster than
regulators are able to address them. The inability of regulators to effectively police financial markets,
coupled with the ability for uncertainty to spread quickly, allows for market failures to be amplified into
panics. Financial market complexity increases uncertainty, which increases the risk of a panic. Id.
185. Cf. supra Part III.A (discussing ring-fencing’s role in helping to correct market failures).
186. See supra text accompanying note 156.
187. See supra text accompanying note 133.
188. See supra text accompanying notes 179–81.
189. Schwarcz, supra note 144, at 248.
190. Id.
191. Id.
192. Id.
193. See supra text accompanying notes 22–79.
194. Cf. supra text accompanying note 176 (referring to financial firms that have the potential to
generate large externalities as SIFIs). The Dodd-Frank Act delegates to regulators the determination of
which financial firms are systemically important. 12 U.S.C. § 5325 (2012).
195. VICKERS REPORT, supra note 2.
2013] RING-FENCING 97
“subsidiarization,” meaning that ring-fencing retail banking operations196
would help ensure that “if a large bank gets into trouble then the damage
could be more easily contained and resolved, protecting depositors and
taxpayers, and thereby preventing or inhibiting the kind of contagion that
leads to widespread systemic instability and the kind of political pressure
that leads to ‘too-big-to-fail’ policies.”197
IV. COSTS AND BENEFITS
The analysis so far has shown that ring-fencing can help to correct
market failures, thereby increasing efficiency and protecting against
systemic risk, by reducing information asymmetry, safeguarding deposit-
taking functions of banks, and limiting the ability of firms to engage in
risky behavior.198 The analysis has also shown that ring-fencing can further
protect against systemic risk by introducing modularity.199
Ring-fencing also has potential costs, however.200 Among other costs,
it can increase the cost of financial services by eliminating the ability of
banks to use low-cost deposits to fund other investments and services.201 It
also can reduce a financial firm’s diversification202 and economy of
scope203 benefits.
This part critiques three types of actual and proposed regulatory uses
of ring-fencing—bank ring-fencing, utility ring-fencing, and the ring-
fencing of SIFIs—in light of their benefits and costs.
196. Recall that the Vickers Report defines retail banking as banking provided for individuals and
small and medium-sized enterprises. See supra text accompanying note 61.
197. Lawrence Baxter, Taking on the Juggernauts, THEPARETOCOMMONS (Apr. 11, 2011),
http://www.theparetocommons.com/2011/04/taking-on-the-juggernauts/.
198. See supra text accompanying notes 186–88.
199. See supra text accompanying notes 189–97.
200. Furthermore, no ring-fencing measure is perfect. For example, despite avoiding Enron’s
bankruptcy, PGE had difficulty accessing short-term capital markets after that bankruptcy. MITCHELL
ET AL., supra note 46, at 14. Furthermore, even when appropriate ring-fencing measures are adopted,
there are still transactional costs. Cf. id. at 7–8 (discussing how ratepayers bear part of the costs of
utility ring-fencing).
201. EU BANK PANEL REPORT, supra note 81, at 99 (discussing the challenges of ring-fencing
banks by separating their commercial banking and trading functions).
202. A financial firm’s business model is built on many dimensions including size, activities,
income model, capital and funding structure, ownership, and corporate structure. See id. at 32–66
(questioning, however, whether there are benefits from diversification in banking).
203. See Lawrence G. Baxter, Betting Big: Value, Caution and Accountability in an Era of Large
Banks and Complex Finance, 31 REV. BANKING & FIN. L. 765, 786–811 (2012) (exploring efficiencies
of scope and scale in big banks and determining that they remain open and very difficult to measure);
infra text accompanying notes 218–19.
98 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
A. BANK RING-FENCING
In the banking context, ring-fencing has been, and is proposed to be,
used primarily to legally deconstruct banks to reallocate and reduce risk by
limiting their ability to engage in risky activities.204 The Glass-Steagall
Act205 represented an actual regulatory use of ring-fencing—and the
Vickers Report206 represents a proposed regulatory use of ring-fencing—
for these purposes.
1. The Glass-Steagall Act
The ring-fencing represented by the Glass-Steagall Act, which legally
deconstructed banks by separating their deposit-taking activities from their
riskier investment banking activities,207 could—and for some banks, may
well—have helped to correct market failures.208 Safeguarding deposits is
arguably beneficial to the public209 and may need regulatory protection
because it appears to suffer from a public-goods problem.210 By legally
isolating deposit-taking banks from liabilities associated with riskier
banking activities, the Glass-Steagall Act helped to safeguard deposits.
The ring-fencing represented by the Glass-Steagall Act could also
have helped correct market failures in the form of information failure
resulting from bounded rationality.211 Bounded rationality can impact
204. Securitization and covered bond transactions raise other ways in which ring-fencing has
been, and is proposed to be, applied to banks. In these transactions, the ring-fencing, discussed supra
text accompanying notes 22–35, is intended, among other things, to achieve the public benefit of
enabling banks to more easily transform their existing inventory of loans into cash from which to make
new loans. Steven L. Schwarcz, The Future of Securitization, 41 CONN. L. REV. 1313, 1315 (2009).
205. See supra text accompanying notes 69–70 (describing the Glass-Steagall Act).
206. See supra text accompanying notes 57–68 (describing the Vickers Report).
207. See supra note 70 and accompanying text.
208. These market failures do not include noncompetitive markets. Recall that banks are neither
monopolies nor oligopolies, and the market for banking activities is competitive. See supra Part III.A.1.
209. Former Federal Reserve Chairman Paul A. Volcker has observed, for example, that banks
perform a critical role in the financial system and in the economy for several reasons, including as
“custodians for the bulk of the liquid savings in the economy.” Paul A. Volcker, Chairman, Bd. of
Governors of the Fed. Reserve Sys., Statement Before the Senate Committee on Banking, Housing, and
Urban Affairs (Apr. 26, 1983), in 69 FED. RES. BULL., May 1983, at 356, 359 [hereinafter Volcker
Statement].
210. See supra text accompanying note 127.
211. Another failure, though not technically a “market failure,” occurs when banks get too big to
manage efficiently. See Baxter, supra note 203, at 818–25 (discussing the problems resulting from the
mergers of large banks during the 2008 financial crisis). Former FDIC Chairperson Sheila Blair
believes, for example, that the big banks are too big to manage centrally and regulate, and they do not
effectively produce shareholder value. She argues that there are management inefficiencies in trying to
centrally manage financial firms that operate so many different business lines, and that smaller, more
specialized firms that focus on core businesses would have better efficiencies, fewer conflicts, and less
2013] RING-FENCING 99
deposit-taking banks by causing bank runs, in which some depositors
panic, causing a grab race that can cause the bank to default.212 By making
deposit-taking banks safer, Glass-Steagall’s ring-fencing would have made
depositors less likely to panic, thereby reducing the risk of bank runs.
Glass-Steagall’s ring-fencing did not appear to have addressed market
failures caused by either agency conflicts213 or, except indirectly,214
responsibility failure. The ring-fencing could have helped to protect against
systemic risk, however, by making deposit-taking banks less risky. It is
unclear, though, if that always represented a net benefit. The dilemma was
that Glass-Steagall’s ring-fencing made deposit-taking banks less risky by
separating the riskier investment banking activities into different legal
entities; and lacking the stability of a traditional banking business, those
different entities would themselves be more likely to fail and thus
systemically risky.
Turning to a cost-benefit analysis, one benefit of Glass-Steagall’s
ring-fencing was that it was a relatively simple rule to implement.215 More
tangibly, Glass-Steagall’s ring-fencing helped to correct several market
failures, thereby safeguarding deposits and reducing the risk of bank
runs.216 The net value of those benefits is unclear, however. In the United
taxpayer risk. Erin Kitzie, CNBC Transcript: Former FDIC Chairman Sheila Blair Speaks with
CNBC’s Scott Wapner Today on “Fast Money Halftime Report,” CNBC (July 25, 2012, 1:34 PM),
http://www.cnbc.com/id/48249787. An incidental benefit of the Glass-Steagall Act is that it would
reduce the size of banks that perform traditional banking activities. The too-big-to-manage problem is
not, however, unique to banks. Peter Fox-Penner, Too Big to Regulate?, THE BASELINE SCENARIO (Jan.
16. 2010), http://baselinescenario.com/2010/01/16/too-big-to-regulate/ (comparing the too-big-to-
regulate problem of utilities—such as the complexity of Enron faced by the U.S. Federal Energy
Regulatory Commission—with that of banks).
212. See supra text accompanying notes 157–59.
213. The agency conflicts of banks do not appear to be significantly different from those of
nonbanks, nor does Glass-Steagall’s ring-fencing appear to address agency conflicts.
214. By reducing the risk of bank runs, the Glass-Steagall Act indirectly would have reduced the
externalities resulting from such a run causing a default, which triggers a system-wide panic. Under the
Volcker Rule, ring-fencing can resolve this responsibility failure by limiting the risky investments that a
bank can make, thereby providing stability not only to individual banks, but also to the system as a
whole by making all banks more stable.
215. See, e.g., Luigi Zingales, Why I Was Won over by Glass-Steagall, FIN. TIMES (June 11,
2012), www.ft.com/intl/cms/s/0/cb3e52be-b08d-11e1-8b36-00144feabdc0.html#axzz2dleGuK8C
(arguing that Glass-Steagall was a simple rule that worked).
216. The safeguarding of deposits not only constitutes a benefit for depositors but also constitutes
a benefit for banks by helping to avoid bank runs. See Gillian G. Garcia, Protecting Bank Deposits, 9
ECON. ISSUES, July 1997, at 1, 2–3, available at http://www.imf.org/external/pubs/ft/issues9/issue9.pdf
(discussing deposit insurance). The safeguarding of deposits can also mean preventing deposited
monies from being used in risky investments. See Ranald Michie & Simon Mollan, British and
American Banking in Historical Perspective: Beware of False Precedents, HIST. & POL’Y. (Dec. 2011),
available at http://www.historyandpolicy.org/papers/policy-paper-128.html (examining the Vickers
100 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
States, at least, government deposit insurance also safeguards deposits and
prevents bank runs; therefore, ring-fencing for those purposes may well
have been duplicative. Similarly, although Glass-Steagall’s ring-fencing
might have reduced systemic risk from traditional banking, it might
inadvertently have increased systemic risk from investment banking.217
It thus is uncertain whether Glass-Steagall’s ring-fencing provided net
benefits. Furthermore, any net benefits would have to be offset by
additional costs, including the possibility that such ring-fencing placed U.S.
banks at a competitive disadvantage with foreign banks.218 Part of this
competitive disadvantage arguably resulted because Glass-Steagall’s ring-
fencing impaired U.S. banks’ economies of scope: that “folding banking in
with insurance, securities, and the like might produce lower costs in
matching sources and uses of funds.”219
It also is unclear whether Glass-Steagall’s ring-fencing, had it applied
during the recent financial crisis, would even have provided net value. One
commentator argued, for example, that “[t]he most telling argument against
a return of Glass-Steagall is that, even if it had been fully in force in 2008,
nothing would have been different.”220 During the crisis, several major U.S.
Report’s proposal for ring-fencing U.K. retail and investment banking and offering alternative solutions
to protect deposits in the United Kingdom). From that perspective, government deposit insurance might
incentivize a bank to use deposited monies in risky investments. See Patricia A. McCoy, The Moral
Hazard Implications of Deposit Insurance: Theory and Evidence, in 5 CURRENT DEVELOPMENTS IN
MONETARY AND FINANCIAL LAW 417, 422–25 (2008) (arguing that deposit insurance can incentivize
banks to take unnecessary risks).
217. See supra text accompanying note 214.
218. See, e.g., WILLIAM JACKSON, CONG. RESEARCH SERV., GLASS-STEAGALL ACT
MODERNIZATION? 9 (1996) (discussing the competitive disadvantage of U.S. banks under the Glass-
Steagall Act); WILLIAM JACKSON, CONG. RESEARCH SERV., GLASS-STEAGALL ACT REFORM 3 (1995)
[hereinafter JACKSON, GLASS-STEAGALL ACT REFORM] (same).
219. WILLIAM JACKSON, CONG. RESEARCH SERV., FINANCIAL MODERNIZATION/GLASS-
STEAGALL ACT ISSUES AND THE FINANCIAL SERVICES ACT OF 1998, H.R. 10 AS PASSED IN THE HOUSE
5 (1998). See also U.S. GEN. ACCOUNTING OFFICE, BANK POWERS: ISSUES RELATED TO REPEAL OF
THE GLASS-STEAGALL ACT 25–27 (1988) (examining economies of scope). Glass-Steagall’s ring-
fencing might have created other costs. Some argue, for example, that its mandated separation caused
bankers and the financial arms of nondepository firms to become competitors. JACKSON, GLASS-
STEAGALL ACT REFORM, supra note 218, at 2. Research has also suggested that Glass-Steagall’s ring-
fencing increased the cost of external finance for corporate investment. See Carlos D. Ramirez, Did
Glass-Steagall Increase the Cost of External Finance for Corporate Investment?: Evidence from Bank
and Insurance Company Affiliations, 59 J. ECON. HIST. 372, 374–83 (1999).
220. Peter J. Wallison, Glass-Steagall Would Have Made No Difference, FIN. TIMES (June 14,
2012), http://ft.com/cms/s/0/5b11f66e-b3ec-11e1-8fea-00144feabdc0.html. Wallison explains that
“[t]he major US commercial banks and investment banks that got into trouble in the 2008 financial
crisis were completely independent of one another. They were unaffiliated before Glass-Steagall was
modified and remained unaffiliated afterwards. So if Glass-Steagall had been fully in force in 2008 it
would have changed nothing.” Id. Wallison’s assessment may not be fair, however, because Glass-
2013] RING-FENCING 101
banks decided, after their own internal studies, not to separate their
traditional (for example, deposit-taking) and investment banking
operations.221 Furthermore, Citigroup commissioned a prominent
management-consulting firm to conduct an independent study of whether it
should be separated into ring-fenced traditional banking and investment
banking entities.222 That study concluded that the separation would be
inefficient.223
2. The Vickers Report
The ring-fencing proposed in the Vickers Report—which (somewhat
like the Glass-Steagall Act) would legally deconstruct banks by separating
traditional retail banking activities (including deposit-taking) from their
riskier investment banking activities224—could help to correct market
failures. Safeguarding retail deposits is beneficial to the public but, because
it (like all deposit-taking225) appears to suffer from a public-goods problem,
it may need regulatory protection. The Vickers Report focuses on the retail
deposit-taking functions of banks.226 By legally separating retail deposit-
taking banking from liabilities associated with riskier banking activities and
from insolvency risks, the Vickers Report’s ring-fencing could help to
safeguard retail deposits.
As with Glass-Steagall’s ring-fencing, the Vickers Report’s ring-
fencing could also help correct information-failure market failures resulting
Steagall’s ring-fencing applied only to affiliated firms. Cf. BARTH & PRABHA, supra note 80, at 24
(“Nor is there clear evidence that . . . separating commercial banking from investment banking would
increase safety. Despite strong separation between the two businesses in the 1980s under the Glass-
Steagall Act, several big banks nevertheless almost failed because of bad loans in Latin America.
Likewise, legions of savings-and-loans failed due to real estate loans.”); Holman W. Jenkins, Jr., Op-
Ed., Sandy Weill Still Doesn’t Have the Answer, WALL ST. J. (July 28, 2012),
http://online.wsj.com/article/SB10000872396390443931404577552913658228058.html (arguing that
restoring the Glass-Steagall Act would change nothing, because governments and banks are too
intertwined due to the size of government debt).
221. Lauren Tara LaCapra, Rick Rothacker & David Henry, Banks Bristle at Breakup Call from
Sandy Weill, REUTERS (July 27, 2012), http://www.reuters.com/article/2012/07/27/banks-weill-
idINL2E8IQF5120120727.
222. See id. (discussing a study performed by Bain & Company at the request of Citigroup’s
Chairman, Sandy Weill).
223. Id. That study is not necessarily dispositive, however, because it is not publicly available for
scrutiny. Id. Citigroup might have had unique circumstances. Moreover, part of the study’s conclusion
was apparently based on tax considerations, id., whereas any adverse tax impact of ring-fencing
presumably could be rendered neutral in a regulatory ring-fencing.
224. See VICKERS REPORT, supra note 2, at 9–12.
225. See supra text accompanying notes 209–10.
226. VICKERS REPORT, supra note 2, at 36–38.
102 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
from bounded rationality, thereby reducing the risk of bank runs.227 Unlike
Glass-Steagall, however, this would constitute a clearer benefit because the
United Kingdom, unlike the United States, lacks government deposit
insurance to safeguard retail deposits and prevent bank runs.228
The Vickers Report’s ring-fencing could also help to protect against
systemic risk by making banks performing traditional retail banking
services less risky. As with Glass-Steagall’s ring-fencing, however, it is
unclear if that will represent a net benefit: those banks would be made less
risky by separating their riskier investment banking activities into different
legal entities that lack the stability of a traditional banking business;
therefore, those different entities would themselves become more likely to
fail and thus systemically risky.229
The Vickers Report’s ring-fencing also purports to protect the banking
function of operating payments systems.230 Like safeguarding deposits,
protecting the operation of payments systems is arguably beneficial to the
public.231 It is unclear, though, if this function needs regulatory protection.
Although operating payments systems is still largely a banking function,
there are an increasing number of “nonbank” private payments systems.
For example, Google Wallet, Square, and iTunes all operate forms of
payments systems without being banks.232
The Vickers Report’s ring-fencing has additional costs and benefits
not dissimilar to those of Glass-Steagall’s ring-fencing. For example,
“Large banks mostly hate the idea [of modularity created by the Vickers
Report] because it inhibits their ability to reorganize, restructure and fund
operations at their will. They claim that the forced structuring imposed by
227. See supra text accompanying notes 211–12.
228. See Asli Demirgüç-Kunt, Baybars Karacaovali & Luc Laeven, Deposit Insurance Around the
World: A Comprehensive Database 75 (World Bank Policy Research Working Paper No. 3628, 2005),
available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=756851 (explaining that in the United
Kingdom the deposit insurance system is government legislated but privately administered and funded;
and that there currently is no public funding for deposit insurance). A full cost-benefit analysis might
also compare the cost of the United Kingdom implementing government deposit insurance.
229. See supra text accompanying note 214.
230. VICKERS REPORT, supra note 2, at 35.
231. Chairman Volcker has observed not only that banks perform a critical role in the financial
system and in the economy as “custodians for the bulk of the liquid savings in the economy” but also as
“operators of the payments system.” Volcker Statement, supra note 209, at 359.
232. See 2011 Evolution of Payments, FIRSTPARTNER,
http://www.ibfsinc.com/downloads/2011_evolution_of_payments_market_map_evaluation.pdf (last
visited Nov. 3, 2013) (mapping trends in payment systems); The History of Money and Payments,
INTUIT, http://payments.intuit.com/history-of-money-and-payments/ (last visited Nov. 3, 2013)
(detailing the evolution of payment systems over time).
2013] RING-FENCING 103
subsidiarization does not match the realities (for which read ‘convenience’)
of daily business operations.”233
Unlike Glass-Steagall, however, the Vickers Report provides its
analysis of the projected costs of implementing its ring-fencing. The
Commission that promulgated that report estimated that its implementation
would directly cost the U.K. banking industry in the range of £4–7 ($6.28–
11) billion per year.234 Above that, it estimated that the cost of lower
economic growth would likely be in the range of £1–3 ($1.57–4.71) billion
per year.235 U.K. banks independently have estimated their implementation
costs to be as much as £10 ($15.71) billion per year.236
These costs, however, should be seen in perspective. An alternative to
ring-fencing, bailing out financial firms that are deemed too big to fail, also
comes with an exorbitant cost—especially if those firms engage in morally
hazardous behavior.237 Ring-fencing can help to mitigate the too-big-to-fail
problem, bringing stability to financial markets. If ring-fencing is
successful, a recent cost-benefit analysis conducted by The Financial Times
in response to the Vickers Report has concluded that the benefits of ring-
fencing should outweigh its costs. The Financial Times compared the
highest official yearly estimate of implementing the Vickers Report, £7
($11) billion,238 with its own estimate of £40 ($62.84) billion as the yearly
cost of enduring financial crises.239 This cost-benefit analysis would
therefore heavily weigh in favor of ring-fencing even if the cost of ring-
fencing were as high as £10 ($15.71) billion per year, the amount
independently estimated by U.K. banks.240
The foregoing balancing assumes, of course, that ring-fencing is
successful: “[T]he costs [of ring-fencing under the Vickers Report] are
clearly only worth paying if the proposals are successful in averting another
crisis.”241 Many are skeptical of the ability of ring-fencing to totally
233. Baxter, supra note 197 (arguing that, nonetheless, “there are a number of broader issues at
stake here, not least of which is protecting the public from the costs of failed bank operations”).
234. Sharlene Goff, Just the Facts: The Vickers Report, FIN. TIMES (Sept. 12, 2011),
http://www.ft.com/intl/cms/s/0/7321c692-dd16-11e0-b4f2-00144feabdc0.html#axzz1t1VC61lr.
235. Id. These estimates do not include the costs of operational changes, such as establishing an
independent board for the bank’s retail arm. Id.
236. Id.
237. Brendan Greeley, The Price of Too Big to Fail, BLOOMBERG BUSINESSWEEK (July 05, 2012),
http://www.businessweek.com/articles/2012-07-05/the-price-of-too-big-to-fail. See also Judge, supra
note 130, at 1267, 1302 (discussing the problem of firms being too-big-to-fail).
238. See Goff, supra note 234.
239. Id.
240. See supra text accompanying note 236.
241. Goff, supra note 234.
104 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
prevent financial crises.242
This cost-benefit analysis does not necessarily include costs resulting
from the difficulty of ring-fenced U.K. banks to compete internationally—a
problem that parallels the problem that Glass-Steagall ring-fenced banks
were arguably at a competitive disadvantage with foreign banks243—and
the impact of that on the U.K. economy. As indicated, that cost has been
estimated to be as high as £3 ($4.71) billion per year.244 Moreover, there is
an intangible cost if, as a result of the Vickers Report ring-fencing, London
loses its attractiveness as a global financial center.245 Nor does the cost-
benefit analysis compare the costs and benefits of partial ring-fencing
measures, such as those recently proposed by the German Ministry of
Finance,246 or the costs and benefits of less invasive alternatives to ring-
242. See, e.g., Adam J. Levitin, In Defense of Bailouts, 99 GEO. L.J. 435, 467 (2011) (“Short of
completely restructuring the financial services marketplace, firewalls will offer incomplete protection at
best.”). At the Fifth Annual Risk Conference of the Federal Reserve Bank of Chicago, Thomas Hoenig,
former President of the Federal Reserve Bank of Kansas City and nominated to be Vice Chair of the
FDIC, responded to the author’s comments on ring-fencing banks by noting, in his experience, the
failure of firewalls. Thomas Hoenig, Comment to author at Fifth Annual Risk Conference of the Federal
Reserve Bank of Chicago (Apr. 10, 2012).
243. See supra text accompanying notes 218–19.
244. See supra text accompanying note 235.
245. Louise Armitstead, George Osborne Reforms Will Devalue British Banks, Analysts Warn,
TELEGRAPH (Feb. 4, 2013), http://www.telegraph.co.uk/finance/newsbysector/banksandfinance/
9847459/George-Osborne-reforms-will-devalue-British-banks-analysts-warn.html. See Levitin, supra
note 242, at 467 (“In a world of competitive global capital markets, attempts to restructure the domestic
financial services industry with an eye to risk compartmentalization could result in firms relocating to
more regulatorily conducive (that is permissive) jurisdictions.”). There also could be costs associated
with enforcing ring-fencing. In a February 2013 speech, Chancellor of the Exchequer George Osborne
announced that the Bank of England will be empowered to break up banks that attempt to circumvent
the ring-fencing implemented under the Vickers Report. Mark Scott, Osborne Promises More
Regulatory Power to Split up British Banks, N.Y. TIMES DEALBOOK (Feb. 04, 2013),
http://dealbook.nytimes.com/2013/02/04/osborne-promises-more-regulatory-power-to-split-up-big-
banks/. This enforcement mechanism has been called “electrifying” the ring-fence. Thomas Pascoe,
George Osborne Misses the Point—Retail Banks, Not Investment Banks, Caused This Crisis,
TELEGRAPH, http://blogs.telegraph.co.uk/finance/thomaspascoe/100022647/george-osbornes-misses-
the-point-retail-banks-not-investment-banks-caused-this-crisis/ (last updated Feb. 04, 2013). Although
Osborne’s proposal to electrify the ring-fence has been met with the criticism that it “could increase the
overall costs of the reform for the [banking] industry,” others observe that, without disincentives, banks
will try to game the rules. Armitstead, supra (internal quotation marks omitted).
246. Press Release, German Fed. Ministry of Fin., German Government Approves Draft Bank-
Separation Law and New Criminal-Law Provisions for the Financial Sector (Feb. 6, 2013), available at
http://www.bundesfinanzministerium.de/Content/EN/Pressemitteilungen/2013/2013-02-06-german-
government-approves-draft-bank-separation-law.html. The German proposal is considered a partial
ring-fencing measure. Although it limits some of the risk to banking activities by requiring many
proprietary trading activities to be placed in a separately capitalized subsidiary, banks are allowed to
continue certain of their risky activities, such as proprietary trading for the purpose of market-making.
Some commentators say this means that “European banks won’t have to ring-fence their risky activities
after all.” George Hay & Dominic Elliott, Living Dangerously Without Ring-Fencing, N.Y. TIMES
2013] RING-FENCING 105
fencing.247
B. UTILITY RING-FENCING
From a cost-benefit standpoint, utility companies represent the easiest
case for ring-fencing. Although utility companies are normally monopolies,
their ring-fencing is not aimed at correcting unfair pricing due to a
monopoly-power market failure.248 Rather, utility companies are ring-
fenced to protect them against internal and external risks, so they can be
assured to be able to continue providing the public with essential utilities
such as power, clean water, and communications.249
The very fact of a utility company being a monopoly effectively
creates a structural mandate for ring-fencing: the utility company should be
protected from risk because it is the only entity in its service area able to
provide its essential services. The benefits of ring-fencing utility companies
that are monopolies250 are therefore likely to exceed the costs.
Contrast monopoly utility companies with banks, which also provide
important public services.251 Even assuming, arguendo, that some banking
services, such as deposit-taking, are essential to the public, the need to ring-
fence banks would not appear to be as strong as the need to ring-fence
utility companies. That is because banks, unlike utility companies, are not
monopolies; indeed, the market for banking services is competitive.252
Therefore, even if some banks become subject to risks that prevent them
from providing their services, other banks would likely be able to provide
DEALBOOK (Jan. 30, 2013), http://dealbook.nytimes.com/2013/01/30/living-dangerously-without-ring-
fencing/.
247. Cf. Alistair Darling, A Crisis Needs a Firewall Not a Ring-fence, FIN. TIMES (Feb. 4, 2013),
http://www.ft.com/cms/s/0/3d164732-6ec7-11e2-9ded-00144feab49a.html?ftcamp=published_
links%2Frss%2Fcompanies_uk%2Ffeed%2F%2Fproduct#axzz2KFdWvkrl (comparing ring-fencing
with requiring higher bank-capital requirements).
248. See supra text accompanying notes 118–19 (explaining why, even though utility companies
are monopolies, ring-fencing’s application to utilities is unrelated to monopoly problems).
249. The regulation of utilities by state public service commissions itself evidences the public-
service nature of the services provided by these utilities. See, e.g., Mission Statement, N.Y. STATE PUB.
SERV. COMM’N, http://www3.dps.ny.gov/W/PSCWeb.nsf/ArticlesByTitle/39108B0E4BEBAB378525
7687006F3A6F?OpenDocument (last visited Nov. 4, 2013) (“The primary mission of the New York
State Department of Public Service is to ensure safe, secure, and reliable access to electric, gas, steam,
telecommunications, and water services for New York State’s residential and business consumers, at
just and reasonable rates. The Department seeks to stimulate innovation, strategic infrastructure
investment, consumer awareness, competitive markets where feasible, and the use of resources in an
efficient and environmentally sound manner.”).
250. This Article does not purport to critique whether utility companies should be monopolies.
251. See supra text accompanying note 209 (discussing the public benefits of deposit-taking).
252. See supra note 117 and accompanying text.
106 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
those services. These differences help to explain why a cost-benefit
analysis for ring-fencing banks needs to be more nuanced and fact-specific
than for ring-fencing utilities.253
C. RING-FENCING OF SIFIS
SIFIs—meaning systemically important financial institutions254—can
include both banks and nonbanks. Ring-fencing can apply to SIFIs in two
ways: by protecting the publicly beneficial activities, if any, performed by
SIFIs, and by protecting against the failure of SIFIs that are so large and
contractually interconnected with other SIFIs (including banks) that their
failure could trigger a systemic collapse.
1. Protecting the Publicly Beneficial Activities Performed by SIFIs
Part IV.A already critiques whether ring-fencing should be used to
protect the publicly beneficial activities performed by SIFIs that are banks.
This Part IV.C.1 therefore focuses on whether ring-fencing should be used
to protect the publicly beneficial activities performed by SIFIs that are not
banks. That inquiry raises a threshold question: What, if anything, is there
about nonbanking finance that is so beneficial to the public that it should be
essential to protect, by regulation if necessary?
In answering this question, it should be noted that, as a result of
disintermediation,255 nonbank SIFIs have begun to perform at least some
services that previously were performed by banks.256 It does not appear,
however, that any of those services are of the type that should justify bank
ring-fencing. Nonbank SIFIs do not take deposits, and at least in the U.S.,
they are legally restricted from doing so.257 Nonbank SIFIs do not operate
253. See supra Part IV.A.
254. See supra text accompanying note 176.
255. Schwarcz, supra note 183, at 626–27.
256. Cf. Volcker Statement, supra note 209, at 360 (suggesting that, as other institutions “take
over” the essential functions of banks, one option for government regulation is to include these
institutions within the regulatory framework).
257. See, e.g., 12 U.S.C. § 26 (2012) (stating that institutions at the national level cannot
“commence the business of banking” without authorization from the Comptroller of the Currency).
Deposit-taking institutions in the United States must receive a license, typically called a charter, which
may be a national charter received from the U.S. Comptroller of the Currency or a state charter received
by the relevant state banking authority. See generally OFFICE OF THE COMPTROLLER OF THE
CURRENCY, CHARTERS: COMPTROLLER’S LICENSING MANUAL (2009), available at
http://www.occ.gov/publications/publications-by-type/licensing-manuals/charters.pdf (detailing the
licensing requirements for a national bank charter); Kenneth E. Scott, In Quest of Reason: The
Licensing Decisions of the Federal Banking Agencies, 42 U. CHI. L. REV. 235 (1975) (discussing the
history and decisionmaking of federal banking agencies).
2013] RING-FENCING 107
payments systems.258 The only traditional banking activity that nonbank
SIFIs are performing is the intermediation of credit, by providing financing
to business.259 Although this activity is beneficial to the public,260 there is
no evidence suggesting that ring-fencing regulation is needed to protect it.
A wide range of nonbank firms engage in disintermediated financing,261
and those that find aspects of ring-fencing desirable as a business matter
are already able to contractually ring-fence themselves.262
2. Protecting Against the Systemic Failure of SIFIs
Another possible use of ring-fencing would be to protect against the
failure of SIFIs that are so large and contractually interconnected with other
SIFIs that their failure could trigger a systemic collapse.263 SIFIs would
thus be required to be ring-fenced not because they perform vital banking
or other activities but, instead, because they pose counterparty risk of
systemic magnitude.
The competing costs and benefits of using ring-fencing to protect
against the systemic failure of SIFIs are highly complex. In the first
instance, such costs and benefits will depend on the ways in which the ring-
fencing is structured.264 The costs of using ring-fencing may also be
somewhat duplicative because ring-fencing is not the only regulatory
solution to this problem; a government could decide, for example, to bail
out failing SIFIs as needed.
Nonetheless, even if its costs are partially duplicative, ring-fencing
might be justified because the cost of a bailout can be exorbitant—not only
the direct bailout cost but also the cost of encouraging SIFIs that view
themselves as too big to fail to engage in morally hazardous behavior.265
An indirect benefit of ring-fencing is that it could help mitigate this too-
258. See supra text accompanying notes 230–32.
259. Schwarcz, supra note 183, at 621, 626–27.
260. Cf. Volcker Statement, supra note 209, at 359 (observing that banks perform a critical role in
the financial system and in the economy by efficiently channeling savings to productive investments—
that is, making loans).
261. Schwarcz, supra note 183, at 626–27.
262. See supra text accompanying note 27 (discussing contractual ring-fencing of SPEs in
securitization transactions). Securitization transactions represent the most dominant form of
disintermediated financing. Schwarcz, supra note 183, at 622.
263. See supra text accompanying notes 189–97.
264. Cf. supra text accompanying notes 193–94 (observing that ring-fencing could insert
modularity into the financial system by using some or all of the tools discussed, including bankruptcy
remoteness, ability to operate on a standalone basis, protection against affiliates, and limitations on
risky activities and investments).
265. See supra text accompanying note 237.
108 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 87:69
big-to-fail problem by protecting against the failure of otherwise too-big-
to-fail SIFIs. On the other hand, some or all of the direct bailout cost might
be able to be privatized, such as through the establishment of a systemic
risk fund.266 But on the other hand still, a privatized systemic risk fund
could be difficult to implement.267
In short, using ring-fencing to protect against the systemic failure of
SIFIs is a complicated subject that requires further study.
V. CONCLUSION
Ring-fencing has been advanced in the United States and abroad as a
regulatory solution to a wide range of financial and business problems. The
term, however, is inconsistently defined and, even within a given
regulatory context, often ill-defined.
Arguing that any definition of a financial regulatory concept should be
rooted pragmatically, this Article begins by analyzing the various real-
world functions of ring-fencing. That analysis shows that when used as a
form of financial regulation, ring-fencing can best be understood as legally
deconstructing a firm in order to more optimally reallocate and reduce risk.
The deconstruction could occur in various ways. For example, the firm
could be made more internally viable, such as by separating risky assets
from the firm, preventing the firm from engaging in risky activities or
investing in risky assets, and ensuring that the firm is able to operate on a
standalone basis even if its affiliates fail. The firm could also be protected
from external risks, such as third-party claims, involuntary bankruptcy, and
affiliate abuse.
Ring-fencing’s reallocation of risk raises important normative
questions about when, and how, it should be used as an economic
regulatory tool. The Article examines and attempts to answer these
questions, taking into account ring-fencing’s potential costs and benefits.
For example, ring-fencing is often considered to help protect certain
publicly beneficial activities that are performed by private-sector firms,
such as utility companies268 and banks.269 From a cost-benefit standpoint,
ring-fencing is highly likely to be appropriate to help protect the publicly
beneficial activities performed by utility companies, such as providing
266. See supra text accompanying notes 176–77.
267. See id.
268. This is the purpose of ring-fencing used to protect essential public utility services.
269. This is the purpose of ring-fencing used under the Glass-Steagall Act and proposed in the
Vickers Report.
2013] RING-FENCING 109
power, clean water, and communications. Not only are those services
essential but the utility company, normally being a monopoly, is the only
entity able to provide the services. Ring-fencing the utility company against
risk helps assure the continuity of those services.
It is less certain, though, that ring-fencing should be used to help
protect other publicly beneficial activities. For example, even if the public
services provided by banks were as important as those provided by public
utilities,270 the need to ring-fence banks would not be as strong as the need
to ring-fence public utilities. That is because the market for banking
services is competitive. If some risky banks become unable to provide
services, other banks should be able to provide substitute services. It
therefore is uncertain whether the benefits of ring-fencing banks would
exceed its costs.
Ring-fencing could also be used to help protect the financial system
itself by mitigating systemic risk and the related too-big-to-fail problem of
large banks and other financial institutions.271 The competing costs and
benefits of using ring-fencing for those purposes, however, would be
highly complex. Not only would they depend, among other things, on the
ways in which the ring-fencing is structured; they also would have to be
compared to the costs and benefits of other regulatory approaches to
mitigating systemic risk.
270. This Article uses the above example solely as an illustration. The Article does not suggest
that the public services provided by banks are as important as those provided by public utilities.
271. This is the purpose of ring-fencing proposed for systemically important financial institutions
under the Dodd-Frank Act.